{"as_of":"2026-08-14T11:30:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:2a797115e036285861da1c475c9bd4f1d0a880b8177319b5708b4a68e9fd573a","coverage":[{"denominator":80,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":80,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-07T14:55:59.575885Z","state":"measured"},{"denominator":81,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":81,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-14T06:32:32.682623+00:00","state":"measured"},{"denominator":1,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":1,"source":"paper_references, paper_reference_links","source_observed_at":"2026-06-30T21:45:39.346685Z","state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"arxiv_reference","source_observed_at":"2026-07-01T14:25:45.885432Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"cited_work":{"arxiv_id":"2505.16875","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2505.16875","snapshot_observed_at":"2026-07-01T14:25:45.885432Z","title":"T2i-conbench: Text-to-image benchmark for continual post-training","venue":null,"work_id":"a0820231-cdb7-4f8f-832c-1e262e3b6422","year":2025},"citing_paper":{"arxiv_id":"2605.14948","last_updated":"2026-05-14T15:20:27Z","snapshot_observed_at":"2026-07-06T23:26:18.733776Z","submitted_at":"2026-05-14T15:20:27Z","title":"ACE-LoRA: Adaptive Orthogonal Decoupling for Continual Image Editing","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-06-30T21:45:39.346685Z"},"links":{"cited_paper":"/paper/2505.16875","citing_paper":"/paper/2605.14948"},"observation_digest":"sha256:db9724ae0e39abfdb384fa4897972c11581161dc48b47814084009e70a9bc50e","observation_id":"bcf2e8cd-7e24-443b-a4cf-19418b885837","resolution":{"observed_at":"2026-07-01T14:25:45.886881Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2505.16875/citation-record","integrity":"/paper/2505.16875/integrity","json":"/paper/2505.16875/citation-record.json","paper":"/paper/2505.16875"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T14:56:10.363587Z","title":"Denton, Seyed Kam- yar Seyed Ghasemipour, Raphael Gontijo Lopes, Burcu Karagol Ayan, Tim Salimans, Jonathan Ho, David J","venue":null,"work_id":"90669ccd-6675-45a1-8c29-33f123c219e3","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:51.517475Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7d574b90b3c42961610784afbe8fcb5b286f2b7c08048b50e2e834e19c24ba3d","observation_id":"d23fab2d-abc6-47a1-b8b3-f6ce19f4cdb5","resolution":{"observed_at":"2026-08-07T14:56:10.451397Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:10.135447Z","title":"High-resolution image synthesis with latent diffusion models.CVPR, 2022","venue":null,"work_id":"dbbaaebd-1e5a-4d8b-a349-8b2e6c79cf12","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:51.645874Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:a84691ffa81505e7b8cd9fede5adb0925023469e49de6bc4f77665793af897b9","observation_id":"3174e729-9c2e-4178-b654-a243587e74ef","resolution":{"observed_at":"2026-08-07T14:56:10.225332Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:09.895562Z","title":"SDXL: improving latent diffusion models for high-resolution image synthesis.ICLR, 2024","venue":null,"work_id":"03b93871-aa0b-4908-a676-a16dc1df8fe2","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:51.759864Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:e9eef7328f4681d3661e69ddb2e96938fb1ccb9453e2e326339bef232f93a27f","observation_id":"18a859ee-763e-4961-862d-6f69615eabe4","resolution":{"observed_at":"2026-08-07T14:56:10.052350Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:09.732804Z","title":"Kwok, Ping Luo, Huchuan Lu, and Zhenguo Li","venue":null,"work_id":"be006982-2186-4019-a542-686dcbbb5f3b","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:51.859765Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:68131b7d1d16cd68dcad6f16b9b56989c3c763373d6d1c68f526818b4b814e65","observation_id":"8660f1ac-1583-4092-8e17-03544e1cf11b","resolution":{"observed_at":"2026-08-07T14:56:09.788926Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:09.441929Z","title":"Pixart- Σ: Weak-to-strong training of diffusion transformer for 4k text-to-image generation.ECCV, 2024","venue":null,"work_id":"7b1d9246-2042-45a1-84fc-0b46ccd794d6","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:51.951720Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:0abec0dab7554123d4f32b4bb91f2e0cfe1762e069a57665b8a4b0a752096d9f","observation_id":"51df0d5d-8c5b-48f3-ae32-9a7a6d160fcc","resolution":{"observed_at":"2026-08-07T14:56:09.605079Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:55:52.031267Z","title":"A survey on personalized content synthesis with diffusion models.arXiv preprint arXiv:2405.05538, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.031267Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:60f9025a705a8e1637de2bf269ab955d0389b8bf5747e92ef43509e4edcd096b","observation_id":"1204c36d-b25a-4ec3-b6dd-64d560252d46","resolution":{"observed_at":"2026-08-07T14:55:52.031267Z","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":"2026-08-07T14:56:09.205580Z","title":"Dream- booth: Fine tuning text-to-image diffusion models for subject-driven generation.CVPR, 2023","venue":null,"work_id":"49b52c59-a4cb-4f53-bc2b-d0cbad6af0cd","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.179068Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:4773cb3dd28c5055506d400949ce70269be4dc6ef2db48bb6ab4f851f3e5469c","observation_id":"e67113ad-e4c6-4abf-ae6a-1ec712148110","resolution":{"observed_at":"2026-08-07T14:56:09.325186Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2505.05573","last_updated":"2025-05-12T17:59:28Z","snapshot_observed_at":"2026-08-10T23:18:57.544703Z","submitted_at":"2025-05-08T18:07:16Z","title":"Prompt to Polyp: Medical Text-Conditioned Image Synthesis with Diffusion Models","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.05573","snapshot_observed_at":"2026-08-07T14:55:52.343614Z","title":"Prompt to polyp: Clinically- aware medical image synthesis with diffusion models.arXiv preprint arXiv:2505.05573, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.343614Z"},"links":{"cited_paper":"/paper/2505.05573","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:5cfa1684ba70c1bcb80f1b5237546f6b21f1caae2e73655d5ecfb9fcda5ddbf5","observation_id":"f62206f5-d5d4-4dba-b22f-cfaa5f4a7b71","resolution":{"observed_at":"2026-08-07T14:55:52.343614Z","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":"2026-08-07T14:56:08.962178Z","title":"Customizing text-to-image diffusion with object viewpoint control.SIGGRAPH Asia, 2024","venue":null,"work_id":"8c1b5cb5-4f1f-4983-880b-ffb524c542d1","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.465210Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:a1cd8f8eaa8c8e729e1707cea315825851970e53e2e3770c867ebf5e753c920b","observation_id":"70af777a-88be-4473-aef1-0bf49e0cccae","resolution":{"observed_at":"2026-08-07T14:56:09.114093Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:08.743278Z","title":"Diffusion based augmenta- tion for captioning and retrieval in cultural heritage.ICCV (Workshops), 2023","venue":null,"work_id":"e9bb0eae-8e8d-4f6d-a5fe-7b4cf5bce1c1","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.581574Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:f4d8c8d95efcdc4d4549b57a85290bb842200cfa4eee01055d47fc23fe604973","observation_id":"04db9b2b-35b7-4b2f-ae13-644731d47b1b","resolution":{"observed_at":"2026-08-07T14:56:08.852828Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:08.432457Z","title":"Parameter-efficient transfer learning for NLP.ICML, 2019","venue":null,"work_id":"c944f290-bafb-41d7-8150-d542ab4d8374","year":2019},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.695868Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:6123337a2fdb844ff169a40dbc4f32c766a944dcf1125e3069930978924a8861","observation_id":"797ef686-d5e5-47ae-a984-0ef6433a30c1","resolution":{"observed_at":"2026-08-07T14:56:08.542457Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:08.184841Z","title":null,"venue":null,"work_id":"123aa57e-3c85-4b0d-9ad7-6933395d4165","year":2021},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:52.828119Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:d4b2ae361fc783f2a57fb3fbf57c17ec0d2760e35be2032fb40d8e136c4d160d","observation_id":"d7e051f0-fd1d-481d-b3a9-d0143fe6eedd","resolution":{"observed_at":"2026-08-07T14:56:08.299598Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:08.047504Z","title":"Towards a unified view of parameter-efficient transfer learning.ICLR, 2022","venue":null,"work_id":"128d229a-5e48-4bd6-88b5-2396b8094198","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.002073Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:19200c8a6fca3692f52c006b486e6c17e3e26292526b2817befdae1fcb32e857","observation_id":"203a61d0-7a10-4e40-acd9-9e05459748bb","resolution":{"observed_at":"2026-08-07T14:56:08.080248Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.906332Z","title":"Fine-tuning large language models for domain adaptation: Exploration of training strategies, scaling, model merging and synergistic capabilities.NPJ Computational Materials, 2025","venue":null,"work_id":"451e1c87-802b-4908-a2ad-45952010bc47","year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.166108Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:81d6c181574e6220216aaf4e79de54e242e1cad015a0f08f15fff97d27a7c971","observation_id":"b0ad766b-96a0-4900-9632-ae40ff9cd1b9","resolution":{"observed_at":"2026-08-07T14:56:07.950358Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.708142Z","title":"Continual diffusion: Continual customization of text-to-image diffusion with c-lora.Trans","venue":null,"work_id":"817ac1fd-8a8e-42b8-ba2a-ebbe536899e8","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.257879Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7b380a6829a7843d19752493589a7c1cf97641e3b1a52be9179bfac2e6ac767e","observation_id":"20273b34-90d6-41dd-8c3a-7d7ceb5ab7e1","resolution":{"observed_at":"2026-08-07T14:56:07.820304Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.513663Z","title":"Continual training of language models for few-shot learning.EMNLP, 2022","venue":null,"work_id":"cf1ee503-6472-43eb-83d2-a2d3ac3f3ed3","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.441512Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:5ee2e86f35bc68840383c7dc5ef8ab5de172f04f3ff81eab5455ae89e8d8683e","observation_id":"173bc682-3ef1-4664-8d37-e0861e1a3fca","resolution":{"observed_at":"2026-08-07T14:56:07.579488Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.357568Z","title":"Continual pre-training of language models.ICLR, 2023","venue":null,"work_id":"4729f40c-07a9-48b1-ba7a-3335f76c0569","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.581403Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:126c124cf141e302ebc94ad6799b7fcbed4906a877a3f136b596dc3650e3a65d","observation_id":"c7d5854b-3eeb-4e0c-9d2a-834a0c23b205","resolution":{"observed_at":"2026-08-07T14:56:07.438455Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.189320Z","title":null,"venue":null,"work_id":"d3474527-e8cc-4b9d-8681-f8404e5c92a8","year":1993},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.698774Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:80aab7b86970847fa30197e529c7c7afaef6e0fa28fd65df14c55b817c9086fb","observation_id":"65cc2fbb-c8e9-4cbb-bf6d-415de74a2568","resolution":{"observed_at":"2026-08-07T14:56:07.284173Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:07.047004Z","title":"Connectionist models of recognition memory: constraints imposed by learning and forgetting functions.Psychological Review, 1990","venue":null,"work_id":"00acadab-8078-42a5-9f7e-2276079ea248","year":1990},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.844128Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7723e78960f79b1964484b09f2913676d6301aeaf03754611b0101d0abe9fd24","observation_id":"08ed5c58-1740-4f6c-80fb-d10a82d5712d","resolution":{"observed_at":"2026-08-07T14:56:07.113214Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:06.941650Z","title":"A comprehensive survey of continual learning: Theory, method and application.IEEE Trans","venue":null,"work_id":"23b00dc7-95e5-43f3-a5ec-a5a80835c65d","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:53.981603Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:90f7b1b29ebd5b9d201aeaa52ab91f5998b3201659d833476c2b313642936f22","observation_id":"5db969ee-bc6f-4a89-91de-6d711c4d6e19","resolution":{"observed_at":"2026-08-07T14:56:06.979547Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1902.10486","last_updated":"2019-06-04T07:59:35Z","snapshot_observed_at":"2026-08-07T08:44:38.959537Z","submitted_at":"2019-02-27T12:34:19Z","title":"On Tiny Episodic Memories in Continual Learning","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1902.10486","snapshot_observed_at":"2026-08-07T14:55:54.153379Z","title":"Dokania, Philip H","venue":null,"work_id":null,"year":1902},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.153379Z"},"links":{"cited_paper":"/paper/1902.10486","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:9c42e282b9a423e0410f60da8186d35f710159dadf37d1fbbe537f0c7b349038","observation_id":"6a210583-9fe8-4f7c-9f49-7084d8350af5","resolution":{"observed_at":"2026-08-07T14:55:54.153379Z","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":"2026-08-07T14:56:06.865522Z","title":"Rusu, Kieran Milan, John Quan, Tiago Ramalho, Agnieszka Grabska-Barwinska, Demis Hassabis, Claudia Clopath, Dharshan Kumaran, and Raia Hadsell","venue":null,"work_id":"2871adbb-8f26-40f4-bf82-598d1850ee71","year":2017},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.310452Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:a9e813eef80d5cdc3dd577da4f30314258d46d30ce1247af769e7c5ddc6e21d0","observation_id":"186b16d5-54ac-46cc-aeb6-da1a12a46152","resolution":{"observed_at":"2026-08-07T14:56:06.884315Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:06.707138Z","title":"Continual learning through synaptic intelligence","venue":null,"work_id":"5d82d266-ad1a-4380-a45c-943125e6f562","year":2017},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.425609Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:a326a7186cbfcc837bab726b76c73a6b2b85e2fb13098dec4edf8f2e429344d8","observation_id":"19c9bca7-ef7b-46e5-ac57-348212fe775b","resolution":{"observed_at":"2026-08-07T14:56:06.784506Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2404.18466","last_updated":"2024-04-29T07:07:58Z","snapshot_observed_at":"2026-08-13T00:19:44.178900Z","submitted_at":"2024-04-29T07:07:58Z","title":"HFT: Half Fine-Tuning for Large Language Models","version":1},"cited_work":{"arxiv_id":"2404.18466","doi":null,"metadata_source":"pith","pith_arxiv_id":"2404.18466","snapshot_observed_at":"2026-08-07T14:56:01.093758Z","title":"HFT: Half Fine-Tuning for Large Language Models","venue":"cs.CL","work_id":"260d33ab-5499-4718-bb6a-9ae8eecafcc6","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.497205Z"},"links":{"cited_paper":"/paper/2404.18466","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:d3121528049eed16e96f43bbbb0ffa5804253d9387ba500644c8a5186e67282d","observation_id":"ec1c5489-2ffa-44b6-abb0-c0e325919784","resolution":{"observed_at":"2026-08-07T14:56:01.233917Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:55:54.564679Z","title":"Mofo: Momentum-filtered optimizer for mitigating forgetting in llm fine-tuning.arXiv preprint arXiv:2407.20999, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.564679Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:34c7f0a7c071c9aa4e0edf2a82901788dc8dd36ae8b941eb779e7cd81a33bd35","observation_id":"7f2d6a24-a591-4565-bdfa-43056489a3a4","resolution":{"observed_at":"2026-08-07T14:55:54.564679Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.16789","last_updated":"2024-11-25T05:27:13Z","snapshot_observed_at":"2026-08-13T00:22:04.546556Z","submitted_at":"2024-04-25T17:38:57Z","title":"Continual Learning of Large Language Models: A Comprehensive Survey","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.16789","snapshot_observed_at":"2026-08-07T14:55:54.664729Z","title":"Continual learning of large language models: A comprehensive survey.arXiv preprint arXiv:2404.16789, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.664729Z"},"links":{"cited_paper":"/paper/2404.16789","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1a4ce98216df5486aac9c3c2d25f64af3dd4082982bf97bbad2b9d584de1d321","observation_id":"545850d2-c155-4d66-b91d-d078b2b5be0b","resolution":{"observed_at":"2026-08-07T14:55:54.664729Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2310.06762","last_updated":"2023-10-10T16:38:49Z","snapshot_observed_at":"2026-08-13T05:52:51.911336Z","submitted_at":"2023-10-10T16:38:49Z","title":"TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.06762","snapshot_observed_at":"2026-08-07T14:55:54.752131Z","title":"Trace: A comprehensive benchmark for continual learning in large language models.arXiv preprint arXiv:2310.06762, 2023","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.752131Z"},"links":{"cited_paper":"/paper/2310.06762","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:65a58b314680942c695e0bccea00c316479f52a9023f90a0127a69b8dd690ffe","observation_id":"c99c9ed8-f4b3-44b9-b189-617ee7da8129","resolution":{"observed_at":"2026-08-07T14:55:54.752131Z","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":"2026-08-07T14:56:06.505232Z","title":"Dick, and Hidenori Tanaka","venue":null,"work_id":"a890a49d-ef68-4251-8c58-1ac1a3cbaf8a","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.840508Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:0b022ff14acdea71d55df93730a2f7e1de198d5243b188f1ea97b42b36aa9482","observation_id":"361e7f10-369e-445b-9629-743c15d2a499","resolution":{"observed_at":"2026-08-07T14:56:06.612763Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":"2501.15857","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T14:56:00.783934Z","title":"Are transformers able to reason by connecting separated knowledge in training data?arXiv preprint arXiv:2501.15857, 2025","venue":null,"work_id":"47d761de-99b1-4fa9-b3ee-165c56209e66","year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.911091Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:d73ecd40d32bf2bb432d126ce45ac1f71628bb41e9a54620798367a380530d33","observation_id":"4821700b-3221-4c3d-90ef-21acc3e8f15f","resolution":{"observed_at":"2026-08-07T14:56:00.875443Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2306.14153","last_updated":"2024-01-16T08:57:11Z","snapshot_observed_at":"2026-08-14T08:05:51.748599Z","submitted_at":"2023-06-25T07:40:39Z","title":"DomainStudio: Fine-Tuning Diffusion Models for Domain-Driven Image Generation using Limited Data","version":4},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2306.14153","snapshot_observed_at":"2026-08-07T14:55:54.983757Z","title":"Domainstudio: Fine-tuning diffusion models for domain-driven image generation using limited data.arXiv preprint arXiv:2306.14153, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:54.983757Z"},"links":{"cited_paper":"/paper/2306.14153","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:33854fa381fa1d4ce2b0a08b4b44c8a1a0f87bc2efa371309d210e150f203d2d","observation_id":"62728e44-29f5-4164-a77e-2742d98220e6","resolution":{"observed_at":"2026-08-07T14:55:54.983757Z","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":"2026-08-07T14:56:06.360585Z","title":"Diffusion models beat gans on image synthesis.NeurIPS, 2021","venue":null,"work_id":"8c38f085-1cfa-4727-9d95-e7cb47cbfe05","year":2021},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.061507Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:9add1a87892a477743432559c430df16c31177f0db917e1a0a0520d5e8572dda","observation_id":"ad4620b4-725f-4116-9270-0144fc422330","resolution":{"observed_at":"2026-08-07T14:56:06.414806Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:06.163408Z","title":"T2i-compbench: A comprehensive benchmark for open-world compositional text-to-image generation.NeurIPS, 2023","venue":null,"work_id":"ada0b0f7-75ef-492c-846f-3c1385d5ba34","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.130564Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7d08f3149eca6dc8a86e1f81ae17353506b9434f0dccc3ad1957743bffc98bd1","observation_id":"000eaafb-19b1-4c42-875d-2a29e066a897","resolution":{"observed_at":"2026-08-07T14:56:06.286548Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:05.972307Z","title":"Geneval: An object-focused framework for evaluating text-to-image alignment.NeurIPS, 2023","venue":null,"work_id":"9e702fab-0bba-4345-ac6f-d5ee209ef13c","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.221542Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:476ffd5ea9c8e97e8e5499520df45a4947f7db2e9449127e2e185d432f538a28","observation_id":"9d260cc7-5d04-4174-a1ff-0f5448c7bb52","resolution":{"observed_at":"2026-08-07T14:56:06.027799Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:05.754528Z","title":"Dreambench++: A human-aligned benchmark for personalized image generation.ICLR, 2025","venue":null,"work_id":"920ae7ed-7817-49ec-93ff-c80573f3d176","year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.297473Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:f7466294d9ea2a6f189160d58e6b4e7e7e789d54477cbda67c95a00317c25a91","observation_id":"6d8d0bdf-9d60-4941-9007-689761b6ae77","resolution":{"observed_at":"2026-08-07T14:56:05.876577Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:05.516619Z","title":null,"venue":null,"work_id":"f912b9e1-eac5-40c0-ab98-1b5ea454dce2","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.387044Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:cabc32e52e8a2245e8125f3ed6073d10e096f8038734ae8502dc8b85c7d36085","observation_id":"85fb8854-68f7-477e-8750-389b0b445d93","resolution":{"observed_at":"2026-08-07T14:56:05.621265Z","resolver_source":"raw_fallback","status":"unresolved"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:05.304719Z","title":"Synthetic data protection: Towards a paradigm change in data regulation?Big Data Soc., 2024","venue":null,"work_id":"8d518ca2-56eb-4456-8ece-9f87cb72b972","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.474381Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:e72f5a44637407e8d8c280890d6dd77d654c6b67419b588e9637be2cdb9a55a8","observation_id":"1513644e-1c6e-4ec1-bdd0-d819d3ada8f8","resolution":{"observed_at":"2026-08-07T14:56:05.398259Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:05.144379Z","title":"Gans trained by a two time-scale update rule converge to a local nash equilibrium.NIPS, 2017","venue":null,"work_id":"b6df659d-7e22-43ec-9197-9ec9e1c1231d","year":2017},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.541875Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:9176990b7f326b0da854d43ab0d1b738bed5c974f64da4d29b17ce289052ddf0","observation_id":"76e9d30c-7d5c-4843-8645-3dac7c7c8cb5","resolution":{"observed_at":"2026-08-07T14:56:05.212148Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:04.891574Z","title":"Belongie, James Hays, Pietro Perona, Deva Ramanan, Piotr Dollár, and C","venue":null,"work_id":"a9d7251a-0e0f-4792-93b3-97ebd7015d40","year":2014},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.601540Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:3cd4fef50d8257825585735a6078071ec9fa404edd0879b082fe61dab4d5f1ad","observation_id":"d05114d9-89ac-46bc-a50d-30b0030a0b4c","resolution":{"observed_at":"2026-08-07T14:56:05.011450Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2501.02189","last_updated":"2025-04-06T03:12:51Z","snapshot_observed_at":"2026-08-13T18:28:46.913210Z","submitted_at":"2025-01-04T04:59:33Z","title":"A Survey of State of the Art Large Vision Language Models: Alignment, Benchmark, Evaluations and Challenges","version":6},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.02189","snapshot_observed_at":"2026-08-07T14:55:55.660939Z","title":"A survey of state of the art large vision language models: Alignment, benchmark, evaluations and challenges.arXiv preprint arXiv:2501.02189, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.660939Z"},"links":{"cited_paper":"/paper/2501.02189","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1c2480e11c2fed8e1be26cc63e0fa47de09b94f125074239385eaee1dcd6fe8f","observation_id":"380a9616-e18a-472a-be36-ed890e532933","resolution":{"observed_at":"2026-08-07T14:55:55.660939Z","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":"2026-08-07T14:56:04.712021Z","title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models.ICLR, 2024","venue":null,"work_id":"58da35c7-013e-4b0d-9514-487249f37734","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.742522Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:d2e43dc7c38ae6080d2091290b31d1c3d50d2a8c7baed5bf5da0fc63af4a5b49","observation_id":"eaec64d7-927b-435a-a8a9-a0f65bc980e6","resolution":{"observed_at":"2026-08-07T14:56:04.802881Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:04.402395Z","title":"Robust visual question answering: Datasets, methods, and future challenges.IEEE Trans","venue":null,"work_id":"9e05c231-1af4-4b5e-964a-da86f657e954","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.853593Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1eb59c206f240f87b995acbff074ae07846f5f92c17b65f4b26613c8ef8a092f","observation_id":"59926ea7-b4d5-41d8-a191-fa194fd091f7","resolution":{"observed_at":"2026-08-07T14:56:04.520446Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:04.191099Z","title":"Human preference score: Better aligning text-to-image models with human preference.ICCV, 2023","venue":null,"work_id":"841a417f-108d-4352-a724-eef36f729479","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:55.928507Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:e9ef1d42beeb6aeaa21a69bcb8841e544f4b7a5bae09af0aa8f2392bcfe33210","observation_id":"bddb4821-2d73-4d01-93e8-a3b2897ba1df","resolution":{"observed_at":"2026-08-07T14:56:04.288634Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:04.090803Z","title":"A theory for knowledge transfer in continual learning.CoLLAs, 2022","venue":null,"work_id":"f45edee9-9e3f-478f-9e26-4a22446bd208","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.041200Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1550dc80e5eeea340f7104ccaaf79de54e719ccf37377dc73a030c7748de3af8","observation_id":"f02fd237-be6d-4468-a3e5-f4d405a9cb4c","resolution":{"observed_at":"2026-08-07T14:56:04.098593Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:03.930170Z","title":"Is multi-task learning an upper bound for continual learning?ICASSP, 2023","venue":null,"work_id":"b98f83b2-43a2-46ad-8201-61010cb01354","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.191573Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:a569d2edc5b8a7d49efc67530ab25b73f4d8c54cbf621e9f474e23bd3110626f","observation_id":"20591492-0095-479d-b3b5-9a2cff205e84","resolution":{"observed_at":"2026-08-07T14:56:04.002864Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.08295","last_updated":"2024-08-15T17:50:07Z","snapshot_observed_at":"2026-08-12T23:03:00.839271Z","submitted_at":"2024-08-15T17:50:07Z","title":"SLCA++: Unleash the Power of Sequential Fine-tuning for Continual Learning with Pre-training","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2408.08295","snapshot_observed_at":"2026-08-07T14:55:56.304312Z","title":"Slca++: Unleash the power of sequential fine-tuning for continual learning with pre-training.arXiv preprint arXiv:2408.08295, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.304312Z"},"links":{"cited_paper":"/paper/2408.08295","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:2066881937e205781b22339d911f4dc6d592b657299c6e1d15932dcf4953c962","observation_id":"f096873e-93e4-49bb-88d5-179778eee440","resolution":{"observed_at":"2026-08-07T14:55:56.304312Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2505.09568","last_updated":"2025-05-14T17:11:07Z","snapshot_observed_at":"2026-07-06T21:23:57.084147Z","submitted_at":"2025-05-14T17:11:07Z","title":"BLIP3-o: A Family of Fully Open Unified Multimodal Models-Architecture, Training and Dataset","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2505.09568","snapshot_observed_at":"2026-08-07T14:55:56.404533Z","title":"Blip3-o: A family of fully open unified multimodal models-architecture, training and dataset.arXiv preprint arXiv:2505.09568, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.404533Z"},"links":{"cited_paper":"/paper/2505.09568","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:5ad693fd3feff4614a5378b17127af8195ae599210b4e283aee18c03e75677a5","observation_id":"41223eae-84fe-4858-b359-6799d27d0706","resolution":{"observed_at":"2026-08-07T14:55:56.404533Z","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":"2026-08-07T14:56:03.818078Z","title":"A statistical theory of regularization-based continual learning.ICML, 2024","venue":null,"work_id":"0fc9ecc8-ac0b-46e1-846a-711d20e8a028","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.495903Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1f50daaf0cabacf4488e4267a69afbe8b81811e0fd74fef0a5e8d94a044fffb3","observation_id":"b6c159a8-ed32-4b4c-a0ef-f78fea04467a","resolution":{"observed_at":"2026-08-07T14:56:03.862024Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:03.676797Z","title":"Approximate fisher information matrix to characterise the training of deep neural networks.IEEE Trans","venue":null,"work_id":"91323eab-002f-42d7-a600-00e7f604f1f4","year":2018},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.611965Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:ea113c37753ce1c5a03c58a50c0e80d23f9e905a8f857ada32c4e504bcf8c4bf","observation_id":"79cb6f1d-f567-4b8f-a3ce-074943793bd6","resolution":{"observed_at":"2026-08-07T14:56:03.733323Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:03.541712Z","title":"BERT: pre-training of deep bidirectional transformers for language understanding.NAACL-HLT, 2019","venue":null,"work_id":"57fd9f93-7c20-413b-8bbd-11ed8fb63dca","year":2019},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.664450Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:0fff8568ca67380961da9fbcc6297d395bbf9fe6ed1e3aa2ff387f69c74f1fcb","observation_id":"e4690716-9cd4-407c-9f58-686bca19fe1a","resolution":{"observed_at":"2026-08-07T14:56:03.617050Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:03.378961Z","title":"Orthogonal subspace learning for language model continual learning.EMNLP, 2023","venue":null,"work_id":"2d4c5f91-0c53-47bc-a383-87d11b1927fb","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.714533Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:83b8abc0ffb27f5768788aca16700400860eec37f1f2d49a1a887ac65e21f58b","observation_id":"dc34078c-c284-4710-bd3d-88decc521aa2","resolution":{"observed_at":"2026-08-07T14:56:03.453360Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:03.257676Z","title":"Sutherland","venue":null,"work_id":"516dbbb2-f71a-4f19-b4db-d8f1585b288f","year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.773040Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:df3d0776392fff100b82683dd86db57b84e5b4e7988b481b781e800cf318c98c","observation_id":"d173c5f5-dc6e-4ec7-a750-cb45bf0de7ab","resolution":{"observed_at":"2026-08-07T14:56:03.306570Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.00645","last_updated":"2025-05-18T05:16:11Z","snapshot_observed_at":"2026-08-12T22:33:27.216812Z","submitted_at":"2024-10-01T12:58:37Z","title":"LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learning","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.00645","snapshot_observed_at":"2026-08-07T14:55:56.858393Z","title":"Tsvd: Bridging theory and practice in continual learning with pre-trained models.arXiv preprint arXiv:2410.00645, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.858393Z"},"links":{"cited_paper":"/paper/2410.00645","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:3e84f2cbe4f5bc4db5e245ff1651558767b7b0542b69a8ba388a45cb4017c903","observation_id":"4751c03e-4852-4d28-ae0b-0b6cf8299b16","resolution":{"observed_at":"2026-08-07T14:55:56.858393Z","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":"2026-08-07T14:56:03.110223Z","title":"Create your world: Lifelong text-to-image diffusion.IEEE Trans","venue":null,"work_id":"c64ee17d-bde7-474e-83d1-550968813ce9","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:56.967629Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:5c2aa0363af3d8b41956b5da63647b0a9b52c91ea18d1fbe9cc90ad3c69e48f3","observation_id":"4feba1f3-fd7f-4a8b-b979-c09ec413a74a","resolution":{"observed_at":"2026-08-07T14:56:03.194563Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2410.16719","last_updated":"2025-04-26T06:15:34Z","snapshot_observed_at":"2026-08-12T22:17:54.008484Z","submitted_at":"2024-10-22T05:59:29Z","title":"Progressive Compositionality in Text-to-Image Generative Models","version":2},"cited_work":{"arxiv_id":"2410.16719","doi":null,"metadata_source":"pith","pith_arxiv_id":"2410.16719","snapshot_observed_at":"2026-08-07T14:56:00.487114Z","title":"Progressive Compositionality in Text-to-Image Generative Models","venue":"cs.CV","work_id":"72c42582-b9cc-439e-a719-20aa96777135","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.101623Z"},"links":{"cited_paper":"/paper/2410.16719","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:154196773bddb3d96e19b6000ee104acf18a1493a4d48bba8e869a1d337b6b50","observation_id":"65cffd4e-310a-4597-8452-78c3530817f4","resolution":{"observed_at":"2026-08-07T14:56:00.580119Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2309.15807","last_updated":"2023-09-27T17:30:19Z","snapshot_observed_at":"2026-08-13T10:03:45.960823Z","submitted_at":"2023-09-27T17:30:19Z","title":"Emu: Enhancing Image Generation Models Using Photogenic Needles in a Haystack","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.15807","snapshot_observed_at":"2026-08-07T14:55:57.193568Z","title":"Emu: Enhancing image generation models using photogenic needles in a haystack.arXiv preprint arXiv:2309.15807, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.193568Z"},"links":{"cited_paper":"/paper/2309.15807","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:6b6acd1b8c5d4f6b446c91b003914cc2cd64ef7ffeffe843851191a746e4b5ae","observation_id":"e5795070-caca-4424-b3f2-f0e30768f368","resolution":{"observed_at":"2026-08-07T14:55:57.193568Z","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":"2026-08-07T14:56:02.989424Z","title":"U-net: Convolutional networks for biomedical image segmentation.MICCAI, 2015","venue":null,"work_id":"9b8de7dc-fbcc-4777-9051-22b5d3479429","year":2015},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.309777Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:f47770f17b4c70537f8aaf266dbc0b0071e086edccc346b7271a664d91df4cf0","observation_id":"93e6205a-ce89-4085-8d55-6a02a9a8db53","resolution":{"observed_at":"2026-08-07T14:56:03.036965Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.846903Z","title":"Learning transferable visual models from natural language supervision.ICML, 2021","venue":null,"work_id":"5e8f9802-64a0-49c9-a607-81c1f31559e9","year":2021},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.432683Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:3b6954f7ed993a6f43015874e0e94e1ccd33c80bcffe92b68686fce4c0ebddb1","observation_id":"abe635c4-395a-4254-9cd9-8fd9e9a24186","resolution":{"observed_at":"2026-08-07T14:56:02.925943Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.729181Z","title":"Improving image captioning with better use of captions","venue":null,"work_id":"a98d8c9a-d542-48ff-8758-4504cfac6d56","year":2020},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.534004Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7eee06117f8fff0e14240e05b8789ed5410f443e99c3f06ca3789f606b927660","observation_id":"8c5846e6-16ad-4093-ae14-24dc7a801559","resolution":{"observed_at":"2026-08-07T14:56:02.781614Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.04295","last_updated":"2024-12-15T13:53:52Z","snapshot_observed_at":"2026-08-12T23:48:31.302741Z","submitted_at":"2024-06-06T17:39:09Z","title":"Everything to the Synthetic: Diffusion-driven Test-time Adaptation via Synthetic-Domain Alignment","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.04295","snapshot_observed_at":"2026-08-07T14:55:57.615254Z","title":"Everything to the synthetic: Diffusion-driven test-time adaptation via synthetic-domain alignment.arXiv preprint arXiv:2406.04295, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.615254Z"},"links":{"cited_paper":"/paper/2406.04295","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:8c4710644fbe380e71723590f59fc4c7ab72619469dc5a6f34ced4e341306cde","observation_id":"c396b7a4-0876-4115-a01a-434955f1c288","resolution":{"observed_at":"2026-08-07T14:55:57.615254Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2304.07429","last_updated":"2023-04-14T23:32:24Z","snapshot_observed_at":"2026-08-13T12:02:34.777924Z","submitted_at":"2023-04-14T23:32:24Z","title":"Identity Encoder for Personalized Diffusion","version":1},"cited_work":{"arxiv_id":"2304.07429","doi":null,"metadata_source":"pith","pith_arxiv_id":"2304.07429","snapshot_observed_at":"2026-08-07T14:56:00.335792Z","title":"Identity Encoder for Personalized Diffusion","venue":"cs.CV","work_id":"b93ac2b1-7b86-4eed-be91-e4e261d2411a","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.728252Z"},"links":{"cited_paper":"/paper/2304.07429","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:dde7edb8b1651d365db881f236138d05041368a5d868aa06e2531f6813196455","observation_id":"26c47138-98a0-4539-a7bc-ed3b37c287de","resolution":{"observed_at":"2026-08-07T14:56:00.382547Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.613380Z","title":"Prompt-to- prompt image editing with cross attention control.ICLR, 2022","venue":null,"work_id":"49e12b97-acb6-4cc1-b241-2f66b8718507","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.843922Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:1252a2cd68eb0f47297197977d836e7b324c515b4b353625d1cbdeba7d4a2c11","observation_id":"148f1bdc-9f15-49b3-bda3-7a37b137f8ab","resolution":{"observed_at":"2026-08-07T14:56:02.658634Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2408.05985","last_updated":"2024-08-12T08:21:04Z","snapshot_observed_at":"2026-08-12T23:05:08.772246Z","submitted_at":"2024-08-12T08:21:04Z","title":"Diffuse-UDA: Addressing Unsupervised Domain Adaptation in Medical Image Segmentation with Appearance and Structure Aligned Diffusion Models","version":1},"cited_work":{"arxiv_id":"2408.05985","doi":null,"metadata_source":"pith","pith_arxiv_id":"2408.05985","snapshot_observed_at":"2026-08-07T14:56:00.142293Z","title":"Diffuse-UDA: Addressing Unsupervised Domain Adaptation in Medical Image Segmentation with Appearance and Structure Aligned Diffusion Models","venue":"cs.CV","work_id":"e1534f9e-1a5c-4d0a-a88d-091a1aa1c455","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:57.994758Z"},"links":{"cited_paper":"/paper/2408.05985","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:b698851a0459a6ee36fe862ff33aa0cb76457dcc91e600b0797546c5e90629eb","observation_id":"aec9f678-4e23-42ab-9cb6-f750344a45bd","resolution":{"observed_at":"2026-08-07T14:56:00.206036Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.419759Z","title":"Diffboost: Enhancing medical image segmentation via text-guided diffusion model.IEEE Transactions on Medical Imaging, 2024","venue":null,"work_id":"8c2a99e2-7674-4aae-8eb4-b547ff4dc535","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.082897Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:b9b6fff63f6af4924283f20f8fbbbcc9ce62a2af333879aefceb5a80814da5f6","observation_id":"73d02449-732d-4cdd-be17-597e7ef674f0","resolution":{"observed_at":"2026-08-07T14:56:02.481446Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.321726Z","title":"Diffportrait3d: Controllable diffusion for zero-shot portrait view synthesis.CVPR, 2024","venue":null,"work_id":"a9b172c5-7f89-44b6-af88-59a30d21914e","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.222437Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:2efd03253e23cc89d80691d3122fde09651817d25bc4b75bf1ad670c6cab1c9f","observation_id":"dd1a52c2-82d4-4827-ad77-2727bf025231","resolution":{"observed_at":"2026-08-07T14:56:02.373784Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2411.10369","last_updated":"2024-11-15T17:19:18Z","snapshot_observed_at":"2026-08-13T17:47:53.908296Z","submitted_at":"2024-11-15T17:19:18Z","title":"Towards High-Fidelity 3D Portrait Generation with Rich Details by Cross-View Prior-Aware Diffusion","version":1},"cited_work":{"arxiv_id":"2411.10369","doi":null,"metadata_source":"pith","pith_arxiv_id":"2411.10369","snapshot_observed_at":"2026-08-07T14:55:59.995919Z","title":"Towards High-Fidelity 3D Portrait Generation with Rich Details by Cross-View Prior-Aware Diffusion","venue":"cs.CV","work_id":"26eda35c-9456-40c0-8e29-6044d523ee78","year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.293323Z"},"links":{"cited_paper":"/paper/2411.10369","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:15d39d4217746de1f2716efc64ecf90b40ccfd407d35051e613c832a81634638","observation_id":"20084761-37c5-427b-91d4-b3346620be5f","resolution":{"observed_at":"2026-08-07T14:56:00.083520Z","resolver_source":"local_arxiv","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1801.01973","last_updated":"2018-06-21T14:54:33Z","snapshot_observed_at":"2026-08-07T19:33:19.839652Z","submitted_at":"2018-01-06T05:44:29Z","title":"A Note on the Inception Score","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1801.01973","snapshot_observed_at":"2026-08-07T14:55:58.393757Z","title":"A note on the inception score.arXiv preprint arXiv:1801.01973, 2018","venue":null,"work_id":null,"year":2018},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.393757Z"},"links":{"cited_paper":"/paper/1801.01973","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:276cbeb55d863bd2f5b5982551025bf517003d4e8fcb79eb124779df63c2ed71","observation_id":"1949151f-277f-4b49-88c0-3d7e6ed9c1a0","resolution":{"observed_at":"2026-08-07T14:55:58.393757Z","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":"2026-08-07T14:56:02.204658Z","title":"Rethinking the inception architecture for computer vision.CVPR, 2016","venue":null,"work_id":"dc1d231f-bea1-497a-a8e5-d282351b7274","year":2016},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.467808Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:d695831ffaadb2bead91f558a45a4ead1096d2dfb5b37b2e56ba759f14b7fdfb","observation_id":"e739c3e5-5862-48e3-973b-f5ef1c9854e7","resolution":{"observed_at":"2026-08-07T14:56:02.257577Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:02.061650Z","title":"Efros, Eli Shechtman, and Oliver Wang","venue":null,"work_id":"caf85f90-0409-4e00-9269-ac4ed21b7514","year":2018},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.539403Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:c577460cd6b9d92b05ca1fb5260fef926a44fc43db5abef45257cbb30f3d6fe5","observation_id":"6ed43eed-1b59-4073-bece-503dbbb05975","resolution":{"observed_at":"2026-08-07T14:56:02.129482Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:01.912484Z","title":"Emerging properties in self-supervised vision transformers.ICCV, 2021","venue":null,"work_id":"58e01df3-66b0-4108-b957-3a874bb0b95f","year":2021},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.637434Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:0d5b6e673ebdc1bcbf362184ef655f7a3021079efc2667cb87b1066a646f5824","observation_id":"2c6e5f04-5d01-4f3a-83e7-c3f939092d41","resolution":{"observed_at":"2026-08-07T14:56:01.990027Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:01.751314Z","title":"Blip: Bootstrapping language-image pre-training for unified vision-language understanding and generation.ICML, 2022","venue":null,"work_id":"5bec41f3-0130-4f72-b587-d14327f5f3e7","year":2022},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.729885Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:aad3dd42d665b88d9f6145927b4df6272c680d1d7cbab3517a7bcf5d993e5a6d","observation_id":"10d05be9-749f-4ea1-ba86-ef8fe8009100","resolution":{"observed_at":"2026-08-07T14:56:01.828516Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+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":"2026-08-07T14:56:01.565043Z","title":"Imagereward: Learning and evaluating human preferences for text-to-image generation.NeurIPS, 2023","venue":null,"work_id":"6a72601f-3020-4c9c-810a-99ea5d1ce381","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":71,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.849018Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:6be7c68280c83002c3039eea52faf9665ccbcf8ab3f6627c4503381d7c724daa","observation_id":"af00fe6c-ad4f-42f1-a4c8-7853dd2c0a11","resolution":{"observed_at":"2026-08-07T14:56:01.654289Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"2406.04584","last_updated":"2024-06-07T02:12:29Z","snapshot_observed_at":"2026-08-12T23:48:16.050712Z","submitted_at":"2024-06-07T02:12:29Z","title":"CLoG: Benchmarking Continual Learning of Image Generation Models","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2406.04584","snapshot_observed_at":"2026-08-07T14:55:58.931222Z","title":"Clog: Benchmarking continual learning of image generation models.arXiv preprint arXiv:2406.04584, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":72,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:58.931222Z"},"links":{"cited_paper":"/paper/2406.04584","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:e7b1c85299deed10b6e2c3c5118d1cf47f8744799095d51e3d3c8ace5f8c3826","observation_id":"547e8f05-0e97-403a-8a92-94de89e4a92b","resolution":{"observed_at":"2026-08-07T14:55:58.931222Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.19437","last_updated":"2025-02-18T17:26:38Z","snapshot_observed_at":"2026-08-11T01:48:59.557045Z","submitted_at":"2024-12-27T04:03:16Z","title":"DeepSeek-V3 Technical Report","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.19437","snapshot_observed_at":"2026-08-07T14:55:59.037860Z","title":"Deepseek-v3 technical report.arXiv preprint arXiv:2412.19437, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":73,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.037860Z"},"links":{"cited_paper":"/paper/2412.19437","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:208bb95df31567bae3a13aa989fdb5898867edcff233e728467df0215720ca9b","observation_id":"cc794c2a-1090-458f-8f31-9925d481eee4","resolution":{"observed_at":"2026-08-07T14:55:59.037860Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2501.12948","last_updated":"2026-01-04T03:57:36Z","snapshot_observed_at":"2026-08-13T15:58:13.809876Z","submitted_at":"2025-01-22T15:19:35Z","title":"DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2501.12948","snapshot_observed_at":"2026-08-07T14:55:59.082383Z","title":"Deepseek-r1: Incentivizing reasoning capability in llms via reinforcement learning.arXiv preprint arXiv:2501.12948, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":74,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.082383Z"},"links":{"cited_paper":"/paper/2501.12948","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7f909f58ab49c38c009b4d26b0ac3320fba005ce5fccac81ab8bcea35bcd269d","observation_id":"f4949118-8bc3-42e1-8e38-a6bc94b423b7","resolution":{"observed_at":"2026-08-07T14:55:59.082383Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2412.15115","last_updated":"2025-01-03T02:18:21Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-12-19T17:56:09Z","title":"Qwen2.5 Technical Report","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2412.15115","snapshot_observed_at":"2026-08-07T14:55:59.179186Z","title":"Qwen2.5 technical report.arXiv preprint arXiv:2412.15115, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":75,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.179186Z"},"links":{"cited_paper":"/paper/2412.15115","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:453a67ad134517d26789d7b900be0d334ac4f0db6ded1fa0cabc322ce43ae601","observation_id":"74cb58b1-0377-442b-8756-2d682f9ae20e","resolution":{"observed_at":"2026-08-07T14:55:59.179186Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2309.17421","last_updated":"2023-10-11T05:07:37Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2023-09-29T17:34:51Z","title":"The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2309.17421","snapshot_observed_at":"2026-08-07T14:55:59.290217Z","title":"The dawn of lmms: Preliminary explorations with gpt-4v(ision).arXiv preprint arXiv:2309.17421, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":76,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.290217Z"},"links":{"cited_paper":"/paper/2309.17421","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:b1aa5ae67c3f892d1bb832b2b99620c88bb169bf07eb3a6afe1e0615e79866ca","observation_id":"61b360f0-ea25-4bb4-a401-d2d259f58d2c","resolution":{"observed_at":"2026-08-07T14:55:59.290217Z","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":"2026-08-07T14:56:01.402376Z","title":"Kingma and Jimmy Ba","venue":null,"work_id":"73752fbb-c104-4286-a3fc-705b9c252daf","year":2015},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":77,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.364506Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:6fe0d1138fa7562b14d7faca526b1d88cba4d1b7e3340ad41f9db5c5c4358f5d","observation_id":"b058e921-b167-4e20-980e-3ac2278c6f46","resolution":{"observed_at":"2026-08-07T14:56:01.466920Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}},{"citation":{"cited_paper":{"arxiv_id":"1910.02054","last_updated":"2020-05-13T06:45:15Z","snapshot_observed_at":"2026-07-06T08:27:00.558613Z","submitted_at":"2019-10-04T17:29:39Z","title":"ZeRO: Memory Optimizations Toward Training Trillion Parameter Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1910.02054","snapshot_observed_at":"2026-08-07T14:55:59.441095Z","title":"Zero: Memory optimizations toward training trillion parameter models.arXiv preprint arXiv:1910.02054, 2020","venue":null,"work_id":null,"year":1910},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":78,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.441095Z"},"links":{"cited_paper":"/paper/1910.02054","citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:7aec80c44110473b2ff38331b5816a00ffe7b8ba63e4bfc4592e8030396d4f3b","observation_id":"cd6ba267-fb86-4c85-b355-b7f53185af27","resolution":{"observed_at":"2026-08-07T14:55:59.441095Z","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":"2026-08-07T14:55:59.520935Z","title":"Decoupled weight decay regularization.ICLR, 2019","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":79,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.520935Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:9ba8657f357e58ccbfffee4069bf76c7621354f7ad4679f278d6d5a39de5ff4d","observation_id":"039c1cd3-c8dd-4572-a342-66740d4a06d2","resolution":{"observed_at":"2026-08-07T14:55:59.520935Z","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":"4500.29190","doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-07T14:55:59.742895Z","title":"item customization","venue":null,"work_id":"226cc2b1-2f1c-4a38-b499-e2685a15efbb","year":2023},"citing_paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training","version":1},"reference_index":80,"source":"pdf_text","source_observed_at":"2026-08-07T14:55:59.575885Z"},"links":{"citing_paper":"/paper/2505.16875"},"observation_digest":"sha256:0d2a6199e6494ef9e0eaaf8c34d8c623bf796f19eb59c8b55c120e1884bbc048","observation_id":"7a7249be-c7cc-42c2-a2f4-c03b4ca4b1ad","resolution":{"observed_at":"2026-08-07T14:55:59.807812Z","resolver_source":"raw_fallback","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-14T06:32:32.682623+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"state":"measured"}}],"paper":{"arxiv_id":"2505.16875","last_updated":"2025-05-22T16:31:43Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-07T21:47:29.169874Z","submitted_at":"2025-05-22T16:31:43Z","title":"T2I-ConBench: Text-to-Image Benchmark for Continual Post-training"},"reference_resolution":{"displayed":80,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":24,"verified_exact":7,"verified_fuzzy":49},"total_outbound_references":80},"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":"2026-08-14T06:32:32.682623+00:00","source":"crossref"},{"observed_at":"2026-08-14T06:32:18.44784+00:00","source":"retraction_watch"}],"thesis":"As of 14 August 2026, this Paper Citation Record lists 80 of 80 outbound references and 1 inbound Pith citation observation for arXiv:2505.16875."}