{"as_of":"2026-08-13T18:05:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:b3c0e9898c63a493c26bbd60567f313375353495ae8239dc8e24f59121b2ab1d","coverage":[{"denominator":107,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":100,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-12T13:52:52.279944Z","state":"measured"},{"denominator":100,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":100,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-13T06:32:02.005865+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2411.16760/citation-record","integrity":"/paper/2411.16760/integrity","json":"/paper/2411.16760/citation-record.json","paper":"/paper/2411.16760"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.853229Z","title":"Quantifying attention flow in transformers","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.853229Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:21a226bec235038925e210772ddb3db1dd4fc619dd477b650f7f0824cbb0b3c5","observation_id":"02a677d0-0937-4c52-81cc-bc4d49fac8cb","resolution":{"observed_at":"2026-08-12T13:52:51.853229Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.857885Z","title":"AttnLRP: Attention- aware layer-wise relevance propagation for transformers","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.857885Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:8d5cacf48f108a9afa3fe3a8f8d1bfb5d1962ae6df4ae97a13aa71d666b00a66","observation_id":"24a4d438-7637-4084-bf00-034d8d831c23","resolution":{"observed_at":"2026-08-12T13:52:51.857885Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.862025Z","title":"XAI for trans- formers: Better explanations through conservative propaga- tion","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.862025Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:08ba8f04d5db8e7f27c1523f90916a0d5cf231c550daef7e617f2dbe778cf4b8","observation_id":"ad720e36-6845-4cc3-a11c-e232f686945b","resolution":{"observed_at":"2026-08-12T13:52:51.862025Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.866083Z","title":null,"venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.866083Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:c1d9befd1b30c3909eb2bbb10d9ff8bc7554e525cb386f515a0a9a5e6c24718c","observation_id":"704658cd-3cd8-4234-a347-a2e3a967b505","resolution":{"observed_at":"2026-08-12T13:52:51.866083Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.870899Z","title":"Anders, David Neumann, Talmaj Marinc, Wojciech Samek, Klaus-Robert M ¨uller, and Sebastian La- puschkin","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.870899Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:44dc55ae22dc238c8c6c8a92734137e35229c85f6821dc473db8e3faa8cea008","observation_id":"5e15fea8-2633-44c3-ad90-88522fe1552e","resolution":{"observed_at":"2026-08-12T13:52:51.870899Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.875201Z","title":"On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation","venue":null,"work_id":null,"year":2015},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.875201Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:013a097e76579f5a76b12532e1254ef26996bc26f747d69f6496b817a2e8b6ae","observation_id":"8eacee86-ec4e-4ffa-843e-4fe1d32b34e2","resolution":{"observed_at":"2026-08-12T13:52:51.875201Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2106.08254","last_updated":"2022-09-03T14:11:33Z","snapshot_observed_at":"2026-08-13T16:48:25.566954Z","submitted_at":"2021-06-15T16:02:37Z","title":"BEiT: BERT Pre-Training of Image Transformers","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2106.08254","snapshot_observed_at":"2026-08-12T13:52:51.879775Z","title":"Beit: Bert pre-training of image transformers","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.879775Z"},"links":{"cited_paper":"/paper/2106.08254","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:5339951318d275cd504fb48a74e29c69d8ea114226fc9c1b57a81ea77e0c0067","observation_id":"0b498d0d-87c9-44bb-9be9-1f5cf51c4377","resolution":{"observed_at":"2026-08-12T13:52:51.879775Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.884266Z","title":"Tenenbaum, and Boris Katz","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.884266Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:4eccc1c08e3399385b43784d426c550d8b148599cb16100408d930d737015aa0","observation_id":"c94c6326-24f4-4315-9836-d930a226d8e8","resolution":{"observed_at":"2026-08-12T13:52:51.884266Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2109.04236","last_updated":"2022-02-16T20:41:48Z","snapshot_observed_at":"2026-08-13T17:45:57.216566Z","submitted_at":"2021-09-09T12:57:06Z","title":"ECQ$^{\\text{x}}$: Explainability-Driven Quantization for Low-Bit and Sparse DNNs","version":2},"cited_work":{"arxiv_id":"2109.04236","doi":null,"metadata_source":"pith","pith_arxiv_id":"2109.04236","snapshot_observed_at":"2026-08-12T13:52:52.635057Z","title":"ECQ$^{\\text{x}}$: Explainability-Driven Quantization for Low-Bit and Sparse DNNs","venue":"cs.LG","work_id":"09c572b9-411a-41a0-b655-6077d2673e18","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.888072Z"},"links":{"cited_paper":"/paper/2109.04236","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:6a272df81fcffafdc56fa4e8ee56e9d704853a81772df2d626a10cd14fce37be","observation_id":"970626d7-7a36-4292-8b6e-287a2aa833bd","resolution":{"observed_at":"2026-08-12T13:52:52.639538Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2006.07159","last_updated":"2020-06-12T13:17:25Z","snapshot_observed_at":"2026-08-11T03:53:10.061881Z","submitted_at":"2020-06-12T13:17:25Z","title":"Are we done with ImageNet?","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2006.07159","snapshot_observed_at":"2026-08-12T13:52:51.892234Z","title":"H’enaff, Alexander Kolesnikov, Xi- aohua Zhai, and A ¨aron van den Oord","venue":null,"work_id":null,"year":2006},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.892234Z"},"links":{"cited_paper":"/paper/2006.07159","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:5f15bbad52fca268b51a611c6b5503e5ddb8c203e28cad164a43743a999bf2b2","observation_id":"2397797f-95ee-409c-9da9-77d99bd6f978","resolution":{"observed_at":"2026-08-12T13:52:51.892234Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.896423Z","title":"Alabdulmohsin, and Filip Pavetic","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.896423Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:60c22267f8a41484498abb6cef4e93d8cfb6cc9c2c140fb58ae077e20c3400d9","observation_id":"e3f88ac5-2eac-46c4-80e1-4ead6d7b3eb0","resolution":{"observed_at":"2026-08-12T13:52:51.896423Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.900078Z","title":"Layer-wise rel- evance propagation for deep neural network architectures","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.900078Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:2a2b793c4a07114b83f5225d459a31718da031b14bc13d8d61b149395b73c12f","observation_id":"025c460c-2887-4504-9664-36d391603e34","resolution":{"observed_at":"2026-08-12T13:52:51.900078Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.903806Z","title":"De- coupling pixel flipping and occlusion strategy for consistent xai benchmarks","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.903806Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:2432f668c9cdc824e8e62f9034342dace3935e02e02ccfe88665ad51eeef50a9","observation_id":"5b222785-5253-4b01-9c69-f37ec52f802c","resolution":{"observed_at":"2026-08-12T13:52:51.903806Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.908286Z","title":"On iden- tifiability in transformers","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.908286Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:2cd0d78aae91737f5d3ac4949b2543a5a05450eb563d3d38c28b4da98bd2c155","observation_id":"4a24f3ca-29a1-4274-b3e6-4fbf0838a514","resolution":{"observed_at":"2026-08-12T13:52:51.908286Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.912677Z","title":"Emerg- ing properties in self-supervised vision transformers","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.912677Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:b1f0ccf22c82c3ef741a61e66d1982ff6cee8bf766ffc53b1c2fb5c06c29a33f","observation_id":"6776aa32-bc6c-475f-96ef-ef78dc843983","resolution":{"observed_at":"2026-08-12T13:52:51.912677Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.916521Z","title":"Generic attention- model explainability for interpreting bi-modal and encoder- decoder transformers","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.916521Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:7e0f0d726edbd538b66fbcb77f0e36d11e55c269a0504bfea69fcf74b6f90e21","observation_id":"ddce0c03-6c20-4248-a65f-198e715fa9c0","resolution":{"observed_at":"2026-08-12T13:52:51.916521Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.920248Z","title":"Transformer inter- pretability beyond attention visualization","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.920248Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:f8017b672159d909b136c12dbfafdabbfa530a142b953d44e9f18b61d67b09ed","observation_id":"f8692399-c916-4518-a30f-99b88f3ca1ef","resolution":{"observed_at":"2026-08-12T13:52:51.920248Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.01161","last_updated":"2022-06-02T17:24:48Z","snapshot_observed_at":"2026-08-13T15:30:59.884781Z","submitted_at":"2022-06-02T17:24:48Z","title":"Optimizing Relevance Maps of Vision Transformers Improves Robustness","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.01161","snapshot_observed_at":"2026-08-12T13:52:51.924114Z","title":"Optimizing relevance maps of vision transformers improves robustness","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.924114Z"},"links":{"cited_paper":"/paper/2206.01161","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:b659f7d8bbb9180cb80d0f54bd6818b0f2555cd2a6fe68d33ca36bd9a45cf91d","observation_id":"39e90983-6213-4e1f-b8c0-912b6b5640c5","resolution":{"observed_at":"2026-08-12T13:52:51.924114Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.13826","last_updated":"2023-05-31T15:42:00Z","snapshot_observed_at":"2026-08-13T12:54:05.601679Z","submitted_at":"2023-01-31T18:10:38Z","title":"Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.13826","snapshot_observed_at":"2026-08-12T13:52:51.928744Z","title":"Attend-and-excite: Attention-based se- mantic guidance for text-to-image diffusion models","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.928744Z"},"links":{"cited_paper":"/paper/2301.13826","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d61cc88ac9ce29344dd8137007b783123c0a7bf3dac6bcbaddbc750a6c0d848b","observation_id":"d727eaa7-f394-49b3-acf4-cac3160197a0","resolution":{"observed_at":"2026-08-12T13:52:51.928744Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.932958Z","title":"Generative pre- training from pixels","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.932958Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:27f7e63a1f47bbb889f9b089241595d21ff53859cb78cada2dcc91da724788b5","observation_id":"578ce905-bde4-430a-9646-927606be930b","resolution":{"observed_at":"2026-08-12T13:52:51.932958Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.936925Z","title":"Training deep nets with sublinear memory cost, 2016","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.936925Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:f22c9dac2bd4defd15b8f4da6b9c8537d207f6ed8ccb68ae124ac0fdd68b21d1","observation_id":"2b723716-3b73-4618-b95d-539f56cb05e5","resolution":{"observed_at":"2026-08-12T13:52:51.936925Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2206.05282","last_updated":"2023-03-01T20:24:58Z","snapshot_observed_at":"2026-08-13T15:26:54.589030Z","submitted_at":"2022-06-10T07:09:28Z","title":"Learning to Estimate Shapley Values with Vision Transformers","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2206.05282","snapshot_observed_at":"2026-08-12T13:52:51.941443Z","title":"Learning to estimate shapley values with vision transformers","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.941443Z"},"links":{"cited_paper":"/paper/2206.05282","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:cced8644b42f12a1679ab52b4245b326c9407c2c4e40998dcb9474d30b89ac11","observation_id":"1699290d-9d10-44ae-b80f-dd09a81d7b79","resolution":{"observed_at":"2026-08-12T13:52:51.941443Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2301.08110","last_updated":"2025-01-07T17:26:26Z","snapshot_observed_at":"2026-08-13T13:01:50.913549Z","submitted_at":"2023-01-19T15:01:00Z","title":"AtMan: Understanding Transformer Predictions Through Memory Efficient Attention Manipulation","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2301.08110","snapshot_observed_at":"2026-08-12T13:52:51.945694Z","title":"Atman: Understanding transformer predictions through memory efficient attention manipulation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.945694Z"},"links":{"cited_paper":"/paper/2301.08110","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:845f13cb4704c7cfcfa995d1699f6f165831de20bf9cbbf2577eb4054e2270c9","observation_id":"8ca71c4a-63e5-4a23-b29a-7eb491db699d","resolution":{"observed_at":"2026-08-12T13:52:51.945694Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.950011Z","title":"Imagenet: A large-scale hierarchical image database","venue":null,"work_id":null,"year":2009},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.950011Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:174f7b6957437a8f13bd978a8e3da8750569fb9d5224be470804880cc1e37dbc","observation_id":"a9312e87-a6ce-4ded-b116-8dcf867355f0","resolution":{"observed_at":"2026-08-12T13:52:51.950011Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.953996Z","title":"An image is worth 16x16 words: Transformers for image recognition at scale","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.953996Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:fe34d6aa4d1821f10b97e810cbb83359e7fb8870b6182587c2372e556e53eabd","observation_id":"5c10ecef-7884-4a3a-bfe0-eedba5f3958b","resolution":{"observed_at":"2026-08-12T13:52:51.953996Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.958583Z","title":"Use hirescam in- stead of grad-cam for faithful explanations of convolutional neural networks","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.958583Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d8bd9d15fd07da7fa8d788e0f3f2b4625be5865223497a72c1df1e3112e2d99c","observation_id":"875e0658-bc58-4e40-a494-8d1e07302298","resolution":{"observed_at":"2026-08-12T13:52:51.958583Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.963338Z","title":"Explain to not forget: Defending against catas- trophic forgetting with xai","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.963338Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:57e65cab2bed0491401d30738ca4650ab4478444f67409a72ea67334ab9e5b2b","observation_id":"23ca89b2-2cb2-4f50-bb1a-c305bee99d9f","resolution":{"observed_at":"2026-08-12T13:52:51.963338Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.967894Z","title":"Eva: Exploring the limits of masked visual representation learning at scale","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.967894Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:4148a177f49ce2081d61ea98679e1d82e33f096def0e868af70a923c7f2ed8a7","observation_id":"cfe1cc0f-f043-41c3-a09e-ac25b1305f76","resolution":{"observed_at":"2026-08-12T13:52:51.967894Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.11331","last_updated":"2023-03-22T14:10:37Z","snapshot_observed_at":"2026-08-13T12:20:30.418845Z","submitted_at":"2023-03-20T17:59:59Z","title":"EVA-02: A Visual Representation for Neon Genesis","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.11331","snapshot_observed_at":"2026-08-12T13:52:51.972241Z","title":"Eva-02: A visual representation for neon genesis","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.972241Z"},"links":{"cited_paper":"/paper/2303.11331","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:246548964eafec4925489998e476a07566faa777cd859bdacb59cc1f95ecb3b1","observation_id":"2e6d70c8-37f7-4363-b0db-506fdb7b8055","resolution":{"observed_at":"2026-08-12T13:52:51.972241Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.976339Z","title":"Adaptive token sampling for efficient vision transformers","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.976339Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:00dfc85140992b9583a65253143f0b9f57bec9e96e9ddb0d5f8e7cf1ed9d6c55","observation_id":"0bbc3ae8-7ced-48de-90da-72a6a19a7e5f","resolution":{"observed_at":"2026-08-12T13:52:51.976339Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.981676Z","title":"Craft: Concept recursive activation factoriza- tion for explainability","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.981676Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:05f504dc1180e19a1f825a27845a926a9f977eaaa76274361549683d2157fcb4","observation_id":"c2c2ad66-6890-47fc-a783-431d23e84ca2","resolution":{"observed_at":"2026-08-12T13:52:51.981676Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.986191Z","title":"G ´allego, and Marta R","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.986191Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:72869337efd05b7c3218d499dd798cec0c14fd7b1247e7d156cd77c9ce562aa4","observation_id":"83f0af78-dae4-44de-b7a1-d6750c4cb06f","resolution":{"observed_at":"2026-08-12T13:52:51.986191Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2008.02312","last_updated":"2020-08-19T06:04:28Z","snapshot_observed_at":"2026-08-10T01:49:21.149915Z","submitted_at":"2020-08-05T18:42:33Z","title":"Axiom-based Grad-CAM: Towards Accurate Visualization and Explanation of CNNs","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2008.02312","snapshot_observed_at":"2026-08-12T13:52:51.989964Z","title":"Axiom-based grad-cam: To- wards accurate visualization and explanation of cnns","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.989964Z"},"links":{"cited_paper":"/paper/2008.02312","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:242551033d6c14c399e9130423465c6283b86c3254a638138ffd9d502808c276","observation_id":"75696c03-af75-4169-8cf2-705f0d566cda","resolution":{"observed_at":"2026-08-12T13:52:51.989964Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.994372Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.994372Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:79e641d9b2b9332528d966bd714b38778efe6f6ef5c041b3139f7c9b41554514","observation_id":"3dd7ec72-5d3b-4f2c-adab-8fb50bcae0ce","resolution":{"observed_at":"2026-08-12T13:52:51.994372Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:51.998492Z","title":"Self-attention at- tribution: Interpreting information interactions inside trans- former","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:51.998492Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:de6b0262804c6c94f2a5626ecb8e45957e0dcdc4b9b0cfe61a3d7f6940993739","observation_id":"01aaec01-54fb-4374-9fea-89efaa3ad1e9","resolution":{"observed_at":"2026-08-12T13:52:51.998492Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.002609Z","title":"Benchmarking neu- ral network robustness to common corruptions and perturba- tions","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.002609Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:94e95dc36e44047db28bc39dc7d5589e7651e32a2020da3a926b3c8fb049894e","observation_id":"c300bca7-4eed-440d-8c36-88aa882ba19f","resolution":{"observed_at":"2026-08-12T13:52:52.002609Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.007347Z","title":"The many faces of robustness: A critical analysis of out-of-distribution generalization","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.007347Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:170013327763a64561944a6a6b4817a871a1fcbdf9561745ce55dda2e9b8b3c4","observation_id":"f4abb6ac-9f51-41af-a555-c054d57f3c44","resolution":{"observed_at":"2026-08-12T13:52:52.007347Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.012053Z","title":"Natural adversarial examples.CVPR,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.012053Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:a84562ec2d7c007179ed3c9998f80eb55bf4e46c85d85f1f623ca3883e4838e2","observation_id":"bdb84a13-a718-44ef-9b28-8932f5edb450","resolution":{"observed_at":"2026-08-12T13:52:52.012053Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2205.13152","last_updated":"2022-05-26T04:59:28Z","snapshot_observed_at":"2026-08-13T15:36:14.526196Z","submitted_at":"2022-05-26T04:59:28Z","title":"Transferable Adversarial Attack based on Integrated Gradients","version":1},"cited_work":{"arxiv_id":"2205.13152","doi":null,"metadata_source":"pith","pith_arxiv_id":"2205.13152","snapshot_observed_at":"2026-08-12T13:52:52.538600Z","title":"Transferable Adversarial Attack based on Integrated Gradients","venue":"cs.LG","work_id":"059cc292-b18a-4e46-a51e-6af5ab7b4673","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.016316Z"},"links":{"cited_paper":"/paper/2205.13152","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d910416f6c8e1ee1258e2760c46d6f9f339d562a6af9531a7ab706a6eed0982d","observation_id":"82bd5a82-b59e-4012-8e7a-27e26e747ebe","resolution":{"observed_at":"2026-08-12T13:52:52.543497Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.316032Z","title":"Ex- plaining convolutional neural networks using softmax gradi- ent layer-wise relevance propagation","venue":null,"work_id":"4b9fcee4-d5ec-4230-b025-ec7f2b5289d3","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.021004Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:6dd852d7e2ab84d9e3bcf147cb6c114ef50f0ce62e1cd8c852e726d8d0a2a2eb","observation_id":"3e53166e-2a11-451c-908a-9027b04df742","resolution":{"observed_at":"2026-08-12T13:52:53.319918Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.301818Z","title":"Layercam: Exploring hierarchical class activation maps for localization","venue":null,"work_id":"d23f6d37-0539-4014-8f77-844485852ba3","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.025457Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:3a584d878133cefa3459978a3cb8fa79a43bb256893da65cb8a1dae42c07cf35","observation_id":"3ca5ceb9-449b-4594-b761-e92c47a1c2fc","resolution":{"observed_at":"2026-08-12T13:52:53.306894Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2308.12964","last_updated":"2023-08-24T17:59:01Z","snapshot_observed_at":"2026-08-13T10:28:08.586978Z","submitted_at":"2023-08-24T17:59:01Z","title":"Dense Text-to-Image Generation with Attention Modulation","version":1},"cited_work":{"arxiv_id":"2308.12964","doi":null,"metadata_source":"pith","pith_arxiv_id":"2308.12964","snapshot_observed_at":"2026-08-12T13:52:52.521918Z","title":"Dense Text-to-Image Generation with Attention Modulation","venue":"cs.CV","work_id":"b06d0e1d-3e14-47f2-a9a7-a18ecef59d2e","year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.030101Z"},"links":{"cited_paper":"/paper/2308.12964","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:bb3c452a295dbe7ca44026952bdcc7811c03fb5a081a5e4c47b391128874a95d","observation_id":"eb6fe22d-b0f7-4617-840f-ba246270e704","resolution":{"observed_at":"2026-08-12T13:52:52.526435Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1611.07270","last_updated":"2016-11-22T12:23:07Z","snapshot_observed_at":"2026-07-06T05:19:39.152493Z","submitted_at":"2016-11-22T12:23:07Z","title":"Investigating the influence of noise and distractors on the interpretation of neural networks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1611.07270","snapshot_observed_at":"2026-08-12T13:52:52.035056Z","title":"Investigating the influence of noise and distractors on the interpretation of neural networks","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.035056Z"},"links":{"cited_paper":"/paper/1611.07270","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d1b17888327c3fc25cb037dab50c615f50b25d5cff97752bb99d9f8fe3c3b867","observation_id":"378adeea-0104-4a4c-bf28-08626ae1df59","resolution":{"observed_at":"2026-08-12T13:52:52.035056Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.288285Z","title":"Attention is not only a weight: Analyzing trans- formers with vector norms","venue":null,"work_id":"bf92ebda-f6bd-4ab0-afdc-93c7cd487c3d","year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.039513Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:89e9cca9d37fed77bec3fbf3e22e5e10aa5af7deb824cd27755f9e54ecb2fcef","observation_id":"4649f801-9857-43f6-aa00-3a0801338116","resolution":{"observed_at":"2026-08-12T13:52:53.292347Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.275103Z","title":"Incorporating Residual and Normalization Layers into Analysis of Masked Language Models","venue":null,"work_id":"305d6333-2559-4a47-a512-e1249e8f4924","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.044200Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:c5699f017aafa495e1a008ddff0dcfd56861853aa8dc8fc27f99e6b810d5de3c","observation_id":"4f6f30b3-edf0-4f95-a8b0-61dcd0c8cd2b","resolution":{"observed_at":"2026-08-12T13:52:53.279569Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.260328Z","title":"Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Doll ´ar, and C","venue":null,"work_id":"3da6d3fc-3f0d-4bcb-ad89-18d85d203e21","year":2014},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.048431Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d07dc7cc5710117508b966f64804d82df86caf51bcec462b36264841dea77d4f","observation_id":"1a55e6d6-512e-4cd8-b6c4-f190ff3b6006","resolution":{"observed_at":"2026-08-12T13:52:53.266769Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2209.11326","last_updated":"2024-01-12T20:19:20Z","snapshot_observed_at":"2026-08-13T14:20:44.430490Z","submitted_at":"2022-09-22T21:40:51Z","title":"Towards Faithful Model Explanation in NLP: A Survey","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2209.11326","snapshot_observed_at":"2026-08-12T13:52:52.053332Z","title":"Towards faithful model explanation in nlp: A survey","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.053332Z"},"links":{"cited_paper":"/paper/2209.11326","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d87c0b8ba2e1a047563bf465e2e9903c6df6221dabd38ecaa579fc341d740766","observation_id":"b8857a91-dad3-44ab-822e-311205461eec","resolution":{"observed_at":"2026-08-12T13:52:52.053332Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.245849Z","title":null,"venue":null,"work_id":"ffd6018b-e791-4299-86a6-6c9edcdbdda3","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.058247Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:0571991ff5756dc3bcd2f1c6cfb8fa57f90eae1236a66f8790563ccb190887b6","observation_id":"7ab46800-1c12-4c86-a15e-aa62a43dc36b","resolution":{"observed_at":"2026-08-12T13:52:53.251212Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.228197Z","title":"SkipPLUS: Skip the first few layers to better explain vision transform- ers","venue":null,"work_id":"fad342b7-3cbc-40be-a86c-f7a4c1d92371","year":2024},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.062593Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:9c55bf11a573cc01b2d46e4e72229345b176a80f8975145b27fedbaf6430724b","observation_id":"a74a3589-55dc-4faa-8700-a9285fe8fb89","resolution":{"observed_at":"2026-08-12T13:52:53.232341Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.200710Z","title":"GlobEnc: Quantifying global token attribution by incorporating the whole encoder layer in transformers","venue":null,"work_id":"d34692ed-be7a-4ebc-9e59-22dce414359a","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.072176Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:015141650f8ee67e37bd0decff862d0773cb1461d507bbf5eb2e24b2cb8c5849","observation_id":"c97a1b7d-4384-47d1-b2e8-99dd96b899fe","resolution":{"observed_at":"2026-08-12T13:52:53.205020Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.188191Z","title":"Modarressi, Hosein Mohebbi, and Mohammad Taher Pilehvar","venue":null,"work_id":"ec68c55c-3fd8-4e95-b4d2-b32bb0752b69","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.076522Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:5103949c99da4bb18c54ef5345a1eead4657d9b65007dba11f26d2bd393f3de3","observation_id":"9095f9eb-3a32-4d17-bd81-c80e28562dee","resolution":{"observed_at":"2026-08-12T13:52:53.192089Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.175246Z","title":"De- compX: Explaining transformers decisions by propagating token decomposition","venue":null,"work_id":"ced69a75-bf7d-443c-ae74-3601e64e370f","year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.080784Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:88c5d874110a99b984731815f48b3606dc8ee7c18fd69b012aeb04b329e184e1","observation_id":"b467daf6-9948-4432-95e1-71c405a2cd54","resolution":{"observed_at":"2026-08-12T13:52:53.180095Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.163510Z","title":"Ex- plaining nonlinear classification decisions with deep taylor decomposition","venue":null,"work_id":"7597a859-47ba-468c-8da7-f01113dc2224","year":2017},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.085231Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:b8da632a5d8aea59c07e81a13b9281710e1f017e54bded2eae21d9ce63ed5cba","observation_id":"e3a5b127-6aa6-4d6f-b7c3-60e92ba8559d","resolution":{"observed_at":"2026-08-12T13:52:53.167471Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.149936Z","title":"Comparing automatic and human evaluation of local explanations for text classification","venue":null,"work_id":"66f036f9-1bc1-4819-9e95-f4d066eee8c8","year":2018},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.088936Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:2ba4b568c70b75e1758d92a188e70ea43883c914bb1991d0df2328cb35236b19","observation_id":"ebd0270c-bba6-46db-9109-fd416a37397e","resolution":{"observed_at":"2026-08-12T13:52:53.155140Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.135851Z","title":"Decompose-and-compose: A composi- tional approach to mitigating spurious correlation","venue":null,"work_id":"f53830f7-2e50-4da7-8a1d-71b62f9c2d61","year":2024},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.092656Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:4d0199eb5ed8b703b4fe1e8ad2ffaa15faf2f97021f9ff2c6e62fe861a906497","observation_id":"7a8c208c-7d28-45c5-895e-fcc6e3069b6e","resolution":{"observed_at":"2026-08-12T13:52:53.140578Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2206.06219","last_updated":"2022-09-27T14:17:31Z","snapshot_observed_at":"2026-08-13T15:24:51.055537Z","submitted_at":"2022-06-13T14:49:17Z","title":"Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure","version":3},"cited_work":{"arxiv_id":"2206.06219","doi":null,"metadata_source":"pith","pith_arxiv_id":"2206.06219","snapshot_observed_at":"2026-08-12T13:52:52.478491Z","title":"Making Sense of Dependence: Efficient Black-box Explanations Using Dependence Measure","venue":"cs.CV","work_id":"80687603-73c1-4c50-a284-9ae7b8009809","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.096382Z"},"links":{"cited_paper":"/paper/2206.06219","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:7da11acb7b2dab575be99eb672b7b6b918bc32dbd4e05d2d4e4b96b89f623fb8","observation_id":"a1eb8359-4fe8-48c1-9ebe-046150b394a7","resolution":{"observed_at":"2026-08-12T13:52:52.483826Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.123473Z","title":"No token left be- hind: Explainability-aided image classification and genera- tion","venue":null,"work_id":"c0723fff-7ff1-40fe-ac91-443b806351dd","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.100281Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:2c50faa00df3cf08f96ee4c9ec0a6b932e3d7c46ceebe6e2acbfdf71d7be49b4","observation_id":"27bb734f-3526-4ef7-be6a-b0a7b32cca2b","resolution":{"observed_at":"2026-08-12T13:52:53.127806Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.110025Z","title":"Parkhi, Andrea Vedaldi, Andrew Zisserman, and C","venue":null,"work_id":"e9870b16-b561-4d05-be73-c7a43f8731fc","year":2012},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.104718Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:00dde06cc180b4994a88b427a4288709892c056fce9513cead4efc7eca91fecb","observation_id":"e85a6e88-a291-49a2-b706-546f5d6813ca","resolution":{"observed_at":"2026-08-12T13:52:53.114507Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2208.06366","last_updated":"2022-10-03T11:47:10Z","snapshot_observed_at":"2026-08-13T14:45:57.387723Z","submitted_at":"2022-08-12T16:48:10Z","title":"BEiT v2: Masked Image Modeling with Vector-Quantized Visual Tokenizers","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2208.06366","snapshot_observed_at":"2026-08-12T13:52:52.108464Z","title":"Beit v2: Masked image modeling with vector-quantized visual tokenizers","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.108464Z"},"links":{"cited_paper":"/paper/2208.06366","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:8854d3c5d7e2b602518d0b65e8dab981e1236624e58743e9cc63ca91e22e0d73","observation_id":"acabef19-02e3-43a6-a201-bf665fe86428","resolution":{"observed_at":"2026-08-12T13:52:52.108464Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1806.07421","last_updated":"2018-09-25T18:16:18Z","snapshot_observed_at":"2026-08-13T09:24:19.606375Z","submitted_at":"2018-06-19T18:41:30Z","title":"RISE: Randomized Input Sampling for Explanation of Black-box Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1806.07421","snapshot_observed_at":"2026-08-12T13:52:52.112291Z","title":"Rise: Random- ized input sampling for explanation of black-box models","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.112291Z"},"links":{"cited_paper":"/paper/1806.07421","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:fbd6e491042d4db9c1799f6a9691da2378ae380ee76714bd618fbf00f7b540b7","observation_id":"c094997f-7b50-4847-af79-4d050c8e4593","resolution":{"observed_at":"2026-08-12T13:52:52.112291Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.097114Z","title":"AttCAT: Explaining transformers via attentive class activation tokens","venue":null,"work_id":"afedb0ba-39ac-4fac-8870-4ec70d6fa7bc","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.117220Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:b7ec38ee8db2691fa4a191a30960e1ad8d231b8e7a9d697b806afc9a511f2d16","observation_id":"8f9c7e47-7e8f-4c77-8cf4-03cd88ad3e17","resolution":{"observed_at":"2026-08-12T13:52:53.101358Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.084148Z","title":"Learning transferable visual models from natural language supervision","venue":null,"work_id":"4e7ef95f-df10-4767-97be-73f7d73648df","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.121652Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d49616da02f07110b0a89296b963186eb3d99b703464e283afb142c3da7a392d","observation_id":"d2f658d0-3cdd-4eff-8bb5-127907ee7e2a","resolution":{"observed_at":"2026-08-12T13:52:53.088550Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1712.06957","last_updated":"2018-05-22T09:19:07Z","snapshot_observed_at":"2026-08-06T21:12:14.834768Z","submitted_at":"2017-12-11T17:46:55Z","title":"MURA: Large Dataset for Abnormality Detection in Musculoskeletal Radiographs","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1712.06957","snapshot_observed_at":"2026-08-12T13:52:52.126492Z","title":"Irvin, Aarti Bagul, Daisy Yi Ding, Tony Duan, Hershel Mehta, Brandon Yang, Kaylie Zhu, Dillon Laird, Robyn L","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.126492Z"},"links":{"cited_paper":"/paper/1712.06957","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:6f83555bde485b54981397f670be227fb3b7aeb8b31ccc57b3274945b1344103","observation_id":"b848166d-9787-4b42-a9d0-62992a9984fc","resolution":{"observed_at":"2026-08-12T13:52:52.126492Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.130635Z","title":"Do imagenet classifiers generalize to im- agenet? In International Conference on Machine Learning,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.130635Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:6ca3d7d20490cc79721a6652ce8a1a8d96c24d07388fa842c3feea3e1069ce42","observation_id":"90b21145-5157-44a6-9cd5-fb3028039fe5","resolution":{"observed_at":"2026-08-12T13:52:52.130635Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.052826Z","title":"why should i trust you?","venue":null,"work_id":"cb2b9ca3-87ae-4e58-8a25-a442d68f5070","year":2016},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.134890Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:be9174649171a2463fdf132b11e96ac9642a522c3d332ce1268ce69d79d7c315","observation_id":"fa97e9ca-041d-4911-86ee-0b5800d37a63","resolution":{"observed_at":"2026-08-12T13:52:53.057580Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.040889Z","title":"Anders, and Klaus-Robert M ¨uller","venue":null,"work_id":"606d6790-b044-4f3a-bb21-2fb106931816","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.139781Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:7b32e29c8d9e4826b77168bd1e0a0704e68caf7aba6ab444bdca3058d02d938e","observation_id":"9577eb9b-2cb8-41e3-a7d6-59301d14d8fe","resolution":{"observed_at":"2026-08-12T13:52:53.045098Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.028182Z","title":"Selvaraju, Michael Cogswell, Abhishek Das, Ramakrishna Vedantam, Devi Parikh, and Dhruv Ba- tra","venue":null,"work_id":"423a3e85-2c4f-485a-afec-09bcd744fea2","year":2017},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.144088Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:9d6f780d271007f1206e733e644a5fe73709cf8b5bf4ed7818e7a535dfc9ee84","observation_id":"f7aa0991-4f7c-459d-b41e-73a22e2f50aa","resolution":{"observed_at":"2026-08-12T13:52:53.032512Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.016143Z","title":"Selvaraju, Stefan Lee, Yilin Shen, Hongxia Jin, Dhruv Batra, and Devi Parikh","venue":null,"work_id":"fcfbaeb4-61b1-4c18-8216-ba9a8707a005","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.148402Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:a39e11caafdd388858a9074303507ec8bf387f1bfb636fb1f9a4e0ec57f3c1e6","observation_id":"898d5c8b-9417-45c8-bf90-caea300cbb74","resolution":{"observed_at":"2026-08-12T13:52:53.020279Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2002.05202","last_updated":"2020-02-12T19:57:13Z","snapshot_observed_at":"2026-08-11T06:21:56.129166Z","submitted_at":"2020-02-12T19:57:13Z","title":"GLU Variants Improve Transformer","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2002.05202","snapshot_observed_at":"2026-08-12T13:52:52.153077Z","title":null,"venue":null,"work_id":null,"year":2002},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.153077Z"},"links":{"cited_paper":"/paper/2002.05202","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:56525c01a94d5bd250cecc502fb68850bb6519ba5bedb1b422e27305709abec8","observation_id":"2f6c931f-7b24-4d5f-872a-c6eed33b7782","resolution":{"observed_at":"2026-08-12T13:52:52.153077Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2302.03318","last_updated":"2023-02-08T15:29:12Z","snapshot_observed_at":"2026-08-13T12:49:37.005639Z","submitted_at":"2023-02-07T08:48:34Z","title":"PAMI: partition input and aggregate outputs for model interpretation","version":2},"cited_work":{"arxiv_id":"2302.03318","doi":null,"metadata_source":"pith","pith_arxiv_id":"2302.03318","snapshot_observed_at":"2026-08-12T13:52:52.411945Z","title":"PAMI: partition input and aggregate outputs for model interpretation","venue":"cs.CV","work_id":"48ecc4c0-8aa6-44be-801c-418abd4f4ac0","year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.157449Z"},"links":{"cited_paper":"/paper/2302.03318","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:a1c832011ebd6eb151b1e11ddaf45b8be6b5790c017f832d51f0aa9a47176f29","observation_id":"34d061d6-cc24-4ac3-9d44-4d07d03f01c8","resolution":{"observed_at":"2026-08-12T13:52:52.418187Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1605.01713","last_updated":"2017-04-11T15:58:48Z","snapshot_observed_at":"2026-08-09T11:23:47.064687Z","submitted_at":"2016-05-05T19:52:32Z","title":"Not Just a Black Box: Learning Important Features Through Propagating Activation Differences","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1605.01713","snapshot_observed_at":"2026-08-12T13:52:52.162026Z","title":"Not just a black box: Learning important features through propagating activation differences","venue":null,"work_id":null,"year":2016},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.162026Z"},"links":{"cited_paper":"/paper/1605.01713","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:8261b72af6ac99935add72f1aaaaf18d1e54e6bd9773e564be923c1249f7b969","observation_id":"a2e5d019-7025-485e-9785-ee8cdc68e6b4","resolution":{"observed_at":"2026-08-12T13:52:52.162026Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:53.003829Z","title":"Learning important features through propagating activation differences","venue":null,"work_id":"3441dc2e-3ad5-4e5c-b1ac-35177e0ac592","year":2017},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.166941Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:264190ff737c29374fefce43b33633d9aa211b8403e5ee052017576e59f6ef8d","observation_id":"35af09fe-5c66-4cc7-b73b-611bd52cf9ee","resolution":{"observed_at":"2026-08-12T13:52:53.007971Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.991999Z","title":"Deep inside convolutional networks: Visualising image clas- sification models and saliency maps","venue":null,"work_id":"b63d7096-f8de-408b-9f8f-f040f3c9b4c5","year":2014},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.170614Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:bd9d7a6bc1d1a82424a8233a8221a296a3323e6419af0c8481bc4f86d12ae229","observation_id":"11f1bb85-b958-41f3-af64-ca0c9cae4040","resolution":{"observed_at":"2026-08-12T13:52:52.996303Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1412.6806","last_updated":"2015-04-13T07:58:17Z","snapshot_observed_at":"2026-08-08T13:18:11.819437Z","submitted_at":"2014-12-21T16:16:37Z","title":"Striving for Simplicity: The All Convolutional Net","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.6806","snapshot_observed_at":"2026-08-12T13:52:52.174481Z","title":"Riedmiller","venue":null,"work_id":null,"year":2014},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.174481Z"},"links":{"cited_paper":"/paper/1412.6806","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:f508268473638940552fd92a2ff0bb3363398870e3ff041fe542d3e680bda796","observation_id":"dac81222-612a-49af-9f7c-c15e8ed176ed","resolution":{"observed_at":"2026-08-12T13:52:52.174481Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.979370Z","title":"Full-gradient represen- tation for neural network visualization","venue":null,"work_id":"21c2bcae-1dfa-4f8a-9fae-f08243bde544","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.178306Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:3d3f4f6bb4a8b3c578ab75481cf5e9f978e1009caf9e5e56a473a5c888827cae","observation_id":"ddc9cd30-b18c-4290-958b-2e612fa14482","resolution":{"observed_at":"2026-08-12T13:52:52.984140Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2303.15389","last_updated":"2023-03-27T17:02:21Z","snapshot_observed_at":"2026-07-06T15:08:34.018146Z","submitted_at":"2023-03-27T17:02:21Z","title":"EVA-CLIP: Improved Training Techniques for CLIP at Scale","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.15389","snapshot_observed_at":"2026-08-12T13:52:52.181963Z","title":"Eva-clip: Improved training techniques for clip at scale","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.181963Z"},"links":{"cited_paper":"/paper/2303.15389","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:a51ac40d75b1f9079a8e6ac3ef0d81a12e663ef218391cd5813383ead19cf6c3","observation_id":"aa686fc4-c339-419e-9303-5ea32bd8515c","resolution":{"observed_at":"2026-08-12T13:52:52.181963Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.967367Z","title":"Axiomatic attribution for deep networks","venue":null,"work_id":"3118e569-977e-44c1-8505-4485d66a7ad6","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.185861Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:b07a4fc20bd59026e71586ee0eab426f4c7308dd4ce389c1802e64b97072b2b9","observation_id":"6bbbffd5-1655-4ec7-bdb6-4149ca435858","resolution":{"observed_at":"2026-08-12T13:52:52.971130Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.943893Z","title":"Imagenet-hard: The hard- est images remaining from a study of the power of zoom and spatial biases in image classification","venue":null,"work_id":"8e054586-928d-4b2d-996a-59daad0ef52f","year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":78,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.192905Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:c201577e61e0ec57a0f232f553f371eb7a03def541a42be317eef62a46b1420a","observation_id":"340ed047-2eae-427b-bf03-2fc941efbef8","resolution":{"observed_at":"2026-08-12T13:52:52.948187Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.932046Z","title":"Tolstikhin, Neil Houlsby, Alexander Kolesnikov, Lu- cas Beyer, Xiaohua Zhai, Thomas Unterthiner, Jessica Yung, Daniel Keysers, Jakob Uszkoreit, Mario Lucic, and Alexey Dosovitskiy","venue":null,"work_id":"a5a63901-2aee-4b5d-ab2e-703e7280cdd1","year":2021},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.196474Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:67576df94fc9c0f848e63b14630e784ab365b28998229b712df82a4d508742b5","observation_id":"c3fbb4b9-639c-4476-a8ed-89d212ad6fcf","resolution":{"observed_at":"2026-08-12T13:52:52.936073Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2012.12877","last_updated":"2021-01-15T15:52:50Z","snapshot_observed_at":"2026-08-09T03:59:14.830524Z","submitted_at":"2020-12-23T18:42:10Z","title":"Training data-efficient image transformers & distillation through attention","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2012.12877","snapshot_observed_at":"2026-08-12T13:52:52.200255Z","title":"Train- ing data-efficient image transformers & distillation through attention","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":80,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.200255Z"},"links":{"cited_paper":"/paper/2012.12877","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:1e2963402ddfc806e307a7e7de7deff84e3e9f39431e24c92082b8d22727d420","observation_id":"cec74a5a-ce47-463f-8a45-a07c5ae045f3","resolution":{"observed_at":"2026-08-12T13:52:52.200255Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.919984Z","title":"Deit iii: Revenge of the vit","venue":null,"work_id":"c2697807-99be-4065-8d51-6264c41e06ca","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":81,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.204352Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:ec557dcb69dcecca3a0e4bcd0db357c8ec7d07b7d4ecd86c6e15f556e51dc5d9","observation_id":"940296d5-b81e-4f4b-84b0-2f35b1a93305","resolution":{"observed_at":"2026-08-12T13:52:52.923824Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.208046Z","title":"Attention is all you need","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":82,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.208046Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:c395601c74c02d94faa7f9f72be17c0103cd91182da237f74570b134df42dcd5","observation_id":"4757cf13-aeab-4d15-a923-49e9cbf0e0a9","resolution":{"observed_at":"2026-08-12T13:52:52.208046Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.901510Z","title":"Analyzing multi-head self-attention: Spe- cialized heads do the heavy lifting, the rest can be pruned","venue":null,"work_id":"b31dc8be-f489-4fd2-88cb-dd24fa5b56dd","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":83,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.211747Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:78fdd7ac6105741c9994eec02992cc2567cb5b59daab51c0072713477c3391ed","observation_id":"3a95145b-0d9e-47ca-8b8c-f4b8d2c2a124","resolution":{"observed_at":"2026-08-12T13:52:52.905839Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.888982Z","title":"Learning robust global representations by penalizing local predictive power","venue":null,"work_id":"4064326b-dc32-43fc-b6e8-f57d014c0291","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":84,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.216730Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:10067799610084f0454bec0585589e5ea10e270b3ad359004a465b899e416527","observation_id":"62083915-9771-4d55-afd3-f9fc4292f1fc","resolution":{"observed_at":"2026-08-12T13:52:52.892852Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.877329Z","title":"Score-cam: Score-weighted visual explanations for convolutional neural networks","venue":null,"work_id":"bac672f4-0757-4430-ad27-68d3c6f40742","year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":85,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.220601Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:e2a29663781dc7f1681171e6b5dc048734e1434120a9ba7233a9545db6dc164f","observation_id":"b371ce4f-4222-4c33-9f0d-bed10d2091f0","resolution":{"observed_at":"2026-08-12T13:52:52.881255Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.865401Z","title":"Beyond explaining: Opportunities and challenges of xai-based model improvement.Inf","venue":null,"work_id":"ce45aa00-9c8b-4699-871b-5d8330e4157b","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":86,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.224179Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:02d04cb794abea7fd94e9a0f4daa2c93cde2d3e435382200e85f69ced054362a","observation_id":"92dc26a1-31db-479d-99c4-0fb6c8703a90","resolution":{"observed_at":"2026-08-12T13:52:52.870032Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.853957Z","title":"Token transformation matters: Towards faithful post-hoc ex- planation for vision transformer","venue":null,"work_id":"35af6b92-46f2-476e-bdf0-7bd3a874dc84","year":2024},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":87,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.227933Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:02ce684a64fc794c28cd7cd73c9b8493785f18096d042f7af8b760e09b096e3e","observation_id":"5b1dce31-5ee9-4279-bb2c-49491244ec19","resolution":{"observed_at":"2026-08-12T13:52:52.858457Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.842183Z","title":"Lyu, and Yu-Wing Tai","venue":null,"work_id":"2c919820-98e3-4d2b-a1a3-808e6ae330e6","year":2020},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":88,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.231728Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:9aa4c2ae4fc73c026b798f312095bf6a66fa49e8f26b8a32aedfec4bc2670b3e","observation_id":"11d9064b-64c2-4c68-8380-e34e4cc3d46c","resolution":{"observed_at":"2026-08-12T13:52:52.846155Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.830043Z","title":"Vit-cx: Causal explanation of vision transformers","venue":null,"work_id":"7b5401e5-6396-4326-a1af-2299f6428d74","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":89,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.236008Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:16855722b2f711c1d962531d987c0ef5393284b8d2314c14fafd220859576af0","observation_id":"9f44bfe5-2c1f-4dcd-ba92-c97650b2e3d5","resolution":{"observed_at":"2026-08-12T13:52:52.834044Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.817967Z","title":null,"venue":null,"work_id":"2ef37c6d-f62b-46c3-ade6-b603d5037763","year":2019},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":90,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.239806Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:90fa82e2299397e6a9751fd8a34c62b020c0cb46cd375f3a1e933c4f6719d3b9","observation_id":"c21d81e7-27ea-4685-a44e-925e505a77cb","resolution":{"observed_at":"2026-08-12T13:52:52.822425Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2405.07992","last_updated":"2024-05-20T16:36:21Z","snapshot_observed_at":"2026-08-13T00:08:38.270310Z","submitted_at":"2024-05-13T17:59:56Z","title":"MambaOut: Do We Really Need Mamba for Vision?","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2405.07992","snapshot_observed_at":"2026-08-12T13:52:52.243617Z","title":"Mambaout: Do we really need mamba for vision? arXiv preprint arXiv:2405.07992,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":91,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.243617Z"},"links":{"cited_paper":"/paper/2405.07992","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:d55056ed3c89782427c1fd32a0577e48be07ccd96404f29b96aac9df7f543899","observation_id":"d372fdef-29c6-4636-a082-7db703779b8e","resolution":{"observed_at":"2026-08-12T13:52:52.243617Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2303.15343","last_updated":"2023-09-27T12:05:41Z","snapshot_observed_at":"2026-07-06T15:08:30.190912Z","submitted_at":"2023-03-27T15:53:01Z","title":"Sigmoid Loss for Language Image Pre-Training","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2303.15343","snapshot_observed_at":"2026-08-12T13:52:52.247875Z","title":"Sigmoid loss for language image pre-training","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":92,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.247875Z"},"links":{"cited_paper":"/paper/2303.15343","citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:bb270881f3c70b79286bb3a19189a53df1457502a24ad1083457b395ac3697db","observation_id":"f0f827e2-7f03-422c-a24e-42cbd5a11e4b","resolution":{"observed_at":"2026-08-12T13:52:52.247875Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.805980Z","title":"Lin, Jonathan Brandt, Xiaohui Shen, and Stan Sclaroff","venue":null,"work_id":"97bcbd6c-227c-4e67-8770-fbca8e232ac3","year":2016},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":93,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.252157Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:cbf3fe7a35feaea013a28251fdeff035a7dad02ea7d26d12247ea4139519bb51","observation_id":"9fbee1a2-e38b-498b-b229-ba77bfa88ad8","resolution":{"observed_at":"2026-08-12T13:52:52.809823Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.795242Z","title":null,"venue":null,"work_id":"5586fb2e-4ae2-4905-9699-729df74a3b86","year":2022},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":94,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.255904Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:6972806878b3f9c9de22775c9b64f21d505d7624d8ef4e39dfbc394f9c389eb4","observation_id":"3575a6ca-6b45-4adc-a34f-1a1a5d0cafba","resolution":{"observed_at":"2026-08-12T13:52:52.798952Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.783895Z","title":null,"venue":null,"work_id":"e1dec915-9151-4bd7-b483-64236de4f714","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":97,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.260238Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:dbadadb4577a419af639753a41254d2551a2f8f828a19de0610547448135f7b2","observation_id":"cc7da1ad-b1b5-432d-b03c-8f8b295223ff","resolution":{"observed_at":"2026-08-12T13:52:52.787777Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.771913Z","title":"Remark 3 (Computational Efficiency)","venue":null,"work_id":"0312d22c-e94c-47aa-9d9f-4e319f451efb","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":98,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.263948Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:9f33185558d437248cedef0950425d7877a13ae7c6d5a3352d4fbe5274ca6583","observation_id":"9daf1987-f6dc-434b-8b1f-ea79d39e6813","resolution":{"observed_at":"2026-08-12T13:52:52.776404Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.760792Z","title":"⊙ f2(x) where [·]cst","venue":null,"work_id":"1b641248-6ac2-454f-9c21-8f9b64eaa81e","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":99,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.268325Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:5020f2288245a0e71ff86957aa9a4dd7914926a6e7dd7e4999e3fe4c7fc27ced","observation_id":"8da33b9b-a08f-40a7-9e07-005a909ff192","resolution":{"observed_at":"2026-08-12T13:52:52.764485Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.748994Z","title":null,"venue":null,"work_id":"40532e48-fcc4-4706-aba1-161381f8cc0b","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":100,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.272244Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:37f0717321073000dfc32c7ec92980b2da15479f5ccc2ae3eb5eb7708ebbf07a","observation_id":"0785114d-70ed-4817-bcde-2e1cb06d99b2","resolution":{"observed_at":"2026-08-12T13:52:52.752911Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.736862Z","title":null,"venue":null,"work_id":"298b706c-49d5-4ffd-b4c4-ab88e7ddf5ea","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":101,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.276354Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:546c2f338a41bb66504067f1a24501c9e74a69a23dd5dd6806f4995a0866b2b1","observation_id":"2842a862-cf4f-42a5-adf8-e011371eba71","resolution":{"observed_at":"2026-08-12T13:52:52.740990Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-12T13:52:52.723908Z","title":"Zebra” and “African Elephant","venue":null,"work_id":"df53291c-848a-411c-9464-2bd3601331ef","year":null},"citing_paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions","version":1},"reference_index":102,"source":"pdf_text","source_observed_at":"2026-08-12T13:52:52.279944Z"},"links":{"citing_paper":"/paper/2411.16760"},"observation_digest":"sha256:a94b00830e7b2da3e72b6ddf6ab389f9c2e430f475084d14985f6f3b934c966f","observation_id":"dafeff7e-fb6d-433e-a585-c204d2e70cff","resolution":{"observed_at":"2026-08-12T13:52:52.728492Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-13T06:32:02.005865+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2411.16760","last_updated":"2024-11-24T15:02:52Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-13T12:04:51.974114Z","submitted_at":"2024-11-24T15:02:52Z","title":"LibraGrad: Balancing Gradient Flow for Universally Better Vision Transformer Attributions"},"reference_resolution":{"displayed":100,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":57,"verified_exact":5,"verified_fuzzy":38},"total_outbound_references":107},"refusal":"A citation records a reference. It does not transfer a finding from one paper to another.","schema":"pith.paper-citation-record.v1","standing_sources":[{"observed_at":"2026-08-13T06:32:02.005865+00:00","source":"crossref"},{"observed_at":"2026-08-13T06:31:53.387327+00:00","source":"retraction_watch"}],"thesis":"As of 13 August 2026, this Paper Citation Record lists 100 of 107 outbound references and 0 inbound Pith citation observations for arXiv:2411.16760."}