{"as_of":"2026-08-09T15:21:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:f231ea268b64ce557c7a391cb6d9ce09de2df8b82f76961d37045214f8aef32d","coverage":[{"denominator":50,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":50,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-03T17:28:44.745087Z","state":"measured"},{"denominator":51,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":51,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-09T06:31:02.800959+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":"2026-07-02T13:55:24.194420Z","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":"2026-07-02T13:56:59.028784Z","state":"measured"}],"external_citation_measurements":[],"inbound":[{"citation":{"cited_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"cited_work":{"arxiv_id":"2512.09446","doi":null,"metadata_source":"arxiv_reference","pith_arxiv_id":"2512.09446","snapshot_observed_at":"2026-07-03T02:17:08.036720Z","title":"Defect-aware hybrid prompt optimization 9 via progressive tuning for zero-shot multi-type anomaly de- tection and segmentation.arXiv preprint arXiv:2512.09446,","venue":null,"work_id":"6ea6cdc4-77c2-45c5-a2fd-c83a1f52937c","year":null},"citing_paper":{"arxiv_id":"2607.01049","last_updated":"2026-07-01T15:11:29Z","snapshot_observed_at":"2026-08-04T06:06:18.855599Z","submitted_at":"2026-07-01T15:11:29Z","title":"GenAU: Language-Grounded Industrial Anomaly Understanding with Vision-Language Models","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-07-02T13:55:24.194420Z"},"links":{"cited_paper":"/paper/2512.09446","citing_paper":"/paper/2607.01049"},"observation_digest":"sha256:709bf33d0e7508b062278b1e70d5222993234b13b1aac3f68c17d651c7d5b615","observation_id":"55d1cc1a-07bf-4d3c-862c-4eca9cf03cb5","resolution":{"observed_at":"2026-07-03T02:17:08.036720Z","resolver_source":"arxiv_id","status":"verified_exact"},"standing_notice":{"events":[],"observation":"No event found in the named queried sources as of 2026-08-09T06:31:02.800959+00:00.","reason":null,"source_receipts":[{"observed_at":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"state":"measured"}}],"links":{"evidence":"/evidence","html":"/paper/2512.09446/citation-record","integrity":"/paper/2512.09446/integrity","json":"/paper/2512.09446/citation-record.json","paper":"/paper/2512.09446"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":null,"snapshot_observed_at":"2026-08-03T17:28:38.949206Z","title":"Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection","venue":null,"work_id":null,"year":2019},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:38.949206Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:e07991bd6d4a73808523fb2063624ff48d7bda7dd2ff728cbee21d33949945e6","observation_id":"1e1445ef-1523-4a76-b7f9-46f20962aaa9","resolution":{"observed_at":"2026-08-03T17:28:38.949206Z","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":"2026-08-03T17:28:39.101215Z","title":"Adaclip: Adapting clip with hybrid learnable prompts for zero-shot anomaly de- tection","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.101215Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:222085fbb83a0a4c524f4b5cbbfd90e1a4f0f08e3b2e9aac788e9aca9e91b256","observation_id":"79e1d942-5a62-4d9e-b6fc-463bf4701d66","resolution":{"observed_at":"2026-08-03T17:28:39.101215Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2305.17382","last_updated":"2023-10-11T07:02:45Z","snapshot_observed_at":"2026-07-06T15:34:16.800916Z","submitted_at":"2023-05-27T06:24:43Z","title":"APRIL-GAN: A Zero-/Few-Shot Anomaly Classification and Segmentation Method for CVPR 2023 VAND Workshop Challenge Tracks 1&2: 1st Place on Zero-shot AD and 4th Place on Few-shot AD","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2305.17382","snapshot_observed_at":"2026-08-03T17:28:39.233436Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.233436Z"},"links":{"cited_paper":"/paper/2305.17382","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:047b12889f90237686855bbadccc08869380be24422a1d8a1602478a13c4a4c6","observation_id":"d5d5101f-f506-4c8f-8d72-246140ce0b2b","resolution":{"observed_at":"2026-08-03T17:28:39.233436Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2005.02357","last_updated":"2021-02-03T16:28:51Z","snapshot_observed_at":"2026-08-09T14:17:32.431497Z","submitted_at":"2020-05-05T17:43:35Z","title":"Sub-Image Anomaly Detection with Deep Pyramid Correspondences","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2005.02357","snapshot_observed_at":"2026-08-03T17:28:39.335652Z","title":"Sub-image anomaly detection with deep pyramid correspondences","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.335652Z"},"links":{"cited_paper":"/paper/2005.02357","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:507133f37d592c2277288e8d2aca8a738cf4164a4b228c4af110b3d523561350","observation_id":"617bea45-2d60-4433-b986-1dfae20b1764","resolution":{"observed_at":"2026-08-03T17:28:39.335652Z","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":"2026-08-03T17:28:39.428767Z","title":"Padim: a patch distribution modeling framework for anomaly detection and localization","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.428767Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:c6c512a5493e6ba0870353315bfef8f2a3c17c2ae5fabbc33acec6ca95dc2103","observation_id":"de20e6df-7d31-49b4-8b87-8fcc4e40a00c","resolution":{"observed_at":"2026-08-03T17:28:39.428767Z","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":"2026-08-03T17:28:39.551809Z","title":"Simclip: Refining image-text alignment with simple prompts for zero-/few-shot anomaly detection","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.551809Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:71197a6fc4c3f92e457b0a44433a5f468ac5d5250c0d3e9f00b368eeff33bf8e","observation_id":"866717b9-40ab-421e-9cb1-abbec01c0fee","resolution":{"observed_at":"2026-08-03T17:28:39.551809Z","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":"2026-08-03T17:28:39.643597Z","title":"Anomaly detection via reverse distillation from one-class embedding","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.643597Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:29c79231cd87ed6eac7fc8089ec846145f3561ce4415200986f27f8e92acf82d","observation_id":"fd08ec47-e340-4a30-aaab-91402ed7fdbb","resolution":{"observed_at":"2026-08-03T17:28:39.643597Z","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":"2026-08-03T17:28:39.800804Z","title":"Generative adversarial networks.Commu- nications of the ACM, 63(11):139–144, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.800804Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:87813b5bac42acce8a8c422d90e957c50c84adaba8921f5ab2709b91fb02ccf1","observation_id":"a00b5621-f477-4372-b2cc-6ec7a76f8c65","resolution":{"observed_at":"2026-08-03T17:28:39.800804Z","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":"2026-08-03T17:28:39.925690Z","title":"Filo: Zero-shot anomaly detection by fine-grained description and high-quality local- ization","venue":null,"work_id":null,"year":2041},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:39.925690Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:216fb5eb5bb0f04d33982c53fe279785d841a9901b549dd3d79357299d117240","observation_id":"dfa0d3fc-081d-4e6d-84c7-35a64cabeb6e","resolution":{"observed_at":"2026-08-03T17:28:39.925690Z","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":"2026-08-03T17:28:40.086815Z","title":"Openclip.If you use this software, please cite it as below, 7,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.086815Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:ba5743c37cecf46ecf211f001aadde5c6d1bffb67d2395d276e59ef0533eb8cb","observation_id":"2a6f0413-3096-4845-b9c4-a461f7abb0c3","resolution":{"observed_at":"2026-08-03T17:28:40.086815Z","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":"2026-08-03T17:28:40.225221Z","title":"Winclip: Zero- /few-shot anomaly classification and segmentation","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.225221Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:456a34cd7b17c738e4b11ba1b0ee632c2a64642200598ed18e66d7b687dc0f89","observation_id":"d9409d02-c158-42a0-8eda-c4941212fce7","resolution":{"observed_at":"2026-08-03T17:28:40.225221Z","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":"2026-08-03T17:28:40.345330Z","title":"Deep learning-based defect detection of metal parts: evaluating current methods in complex condi- tions","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.345330Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:0dde3b48919c37a2485a3e68d3e6239a51c4999bc68c15a4e59069fa4f826a5e","observation_id":"6303fc3e-bc74-4665-8e34-4c2cf4de33bf","resolution":{"observed_at":"2026-08-03T17:28:40.345330Z","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":"2026-08-03T17:28:40.483113Z","title":"How can we know what language models know?Trans- actions of the Association for Computational Linguistics, 8: 423–438, 2020","venue":null,"work_id":null,"year":2020},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.483113Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:2c6b8087a989276a5d45a2e9f74a47b181edbb70ac58909069000261b9733ae1","observation_id":"d2b39e2f-88d9-454f-b9f7-44b267591c2f","resolution":{"observed_at":"2026-08-03T17:28:40.483113Z","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":"2026-08-03T17:28:40.560063Z","title":"Maple: Multi-modal prompt learning","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.560063Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:b9542e4f87992f32d12e8e27593f8c247dccd224e507193291f11db4cbaf03bd","observation_id":"59605650-9336-4260-a5f2-0cc132e12df1","resolution":{"observed_at":"2026-08-03T17:28:40.560063Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1412.6980","last_updated":"2017-01-30T01:27:54Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2014-12-22T13:54:29Z","title":"Adam: A Method for Stochastic Optimization","version":9},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1412.6980","snapshot_observed_at":"2026-08-03T17:28:40.683123Z","title":"Adam: A method for stochastic optimization.arXiv preprint arXiv:1412.6980,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.683123Z"},"links":{"cited_paper":"/paper/1412.6980","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:5014d3e44c7c3373ba7b33aa7179021554da659dfe450f1e607180bc520b7b0f","observation_id":"b5259443-df91-4344-adb8-64b9212f99ef","resolution":{"observed_at":"2026-08-03T17:28:40.683123Z","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":"2026-08-03T17:28:40.799542Z","title":"Auto-encoding vari- ational bayes, 2013","venue":null,"work_id":null,"year":2013},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.799542Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:222f94fcc5000683c761d7ee1073c99a162cb787a1ce2db15f9452067c0c8085","observation_id":"9204adca-4d09-435f-80f5-6ce7d609c7ae","resolution":{"observed_at":"2026-08-03T17:28:40.799542Z","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":"2026-08-03T17:28:40.915067Z","title":"Segment any- thing","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:40.915067Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:4c77492d55e079040afce9b794008a901bc8b61788a70d72877ff27235147b4b","observation_id":"89db07f6-ffef-4f51-bc0b-fcb754421a49","resolution":{"observed_at":"2026-08-03T17:28:40.915067Z","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":"2026-08-03T17:28:41.053086Z","title":"Generalization and network design strate- gies.Connectionism in perspective, 19(143-155):18, 1989","venue":null,"work_id":null,"year":1989},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.053086Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:48f6827d417ec180c9594e3ff1b5410ffa69341922cb6d8758826b7bb7837bc5","observation_id":"03563075-eff1-47ee-bf40-36021911fe41","resolution":{"observed_at":"2026-08-03T17:28:41.053086Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2104.08691","last_updated":"2021-09-02T17:34:41Z","snapshot_observed_at":"2026-08-06T15:24:34.790850Z","submitted_at":"2021-04-18T03:19:26Z","title":"The Power of Scale for Parameter-Efficient Prompt Tuning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2104.08691","snapshot_observed_at":"2026-08-03T17:28:41.160871Z","title":"The power of scale for parameter-efficient prompt tuning.arXiv preprint arXiv:2104.08691, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.160871Z"},"links":{"cited_paper":"/paper/2104.08691","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:01763f1b85b5eb9a8c0e55280790f7e2e188025e218052ca198013c06610ab1b","observation_id":"05277449-beb6-455e-89d0-4f91af441cb9","resolution":{"observed_at":"2026-08-03T17:28:41.160871Z","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":"2026-08-03T17:28:41.261412Z","title":"Clipsam: Clip and sam collaboration for zero-shot anomaly segmentation.Neurocomputing, 618: 129122, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.261412Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:b407efbdabf8d297f619afb0e678ab3eb97937bfa609f3d1e229f5c944a3b4e3","observation_id":"84584541-ff74-4d5f-a475-95cea1c11ef3","resolution":{"observed_at":"2026-08-03T17:28:41.261412Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"1911.02855","last_updated":"2026-07-29T07:55:45Z","snapshot_observed_at":"2026-08-01T23:09:00.410365Z","submitted_at":"2019-11-07T11:14:05Z","title":"Dice Loss for Data-imbalanced NLP Tasks","version":5},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"1911.02855","snapshot_observed_at":"2026-08-03T17:28:41.373771Z","title":"Dice loss for data-imbalanced nlp tasks","venue":null,"work_id":null,"year":1911},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.373771Z"},"links":{"cited_paper":"/paper/1911.02855","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:c5cf2da26df737182b787965b7ef4c0734305bc78aae726cb9ead5bca9e2dfe9","observation_id":"e0b1ab3b-ee41-4a7b-bd29-4505150c5d7b","resolution":{"observed_at":"2026-08-03T17:28:41.373771Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2101.00190","last_updated":"2021-01-01T08:00:36Z","snapshot_observed_at":"2026-07-06T10:29:18.734092Z","submitted_at":"2021-01-01T08:00:36Z","title":"Prefix-Tuning: Optimizing Continuous Prompts for Generation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2101.00190","snapshot_observed_at":"2026-08-03T17:28:41.438414Z","title":"Prefix-tuning: Optimiz- ing continuous prompts for generation.arXiv preprint arXiv:2101.00190, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.438414Z"},"links":{"cited_paper":"/paper/2101.00190","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:cc51252f35812252ed9d41815382228b0ea0c3b3ad056d6c64236ad7153a4556","observation_id":"8501638c-3aae-4462-99c8-74136e091c43","resolution":{"observed_at":"2026-08-03T17:28:41.438414Z","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":"2026-08-03T17:28:41.521244Z","title":"Focal loss for dense object detection","venue":null,"work_id":null,"year":2017},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.521244Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:6d5adef18fc4306a020a6396e004c89a529ed6f190e60ca64a85aaf575cb46bf","observation_id":"68a90c82-a303-4178-888d-6b7bfbba0082","resolution":{"observed_at":"2026-08-03T17:28:41.521244Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2103.11671","last_updated":"2021-03-22T08:57:27Z","snapshot_observed_at":"2026-08-04T10:01:29.161020Z","submitted_at":"2021-03-22T08:57:27Z","title":"Unsupervised Two-Stage Anomaly Detection","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2103.11671","snapshot_observed_at":"2026-08-03T17:28:41.597692Z","title":"Unsu- pervised two-stage anomaly detection.arXiv preprint arXiv:2103.11671, 2021","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.597692Z"},"links":{"cited_paper":"/paper/2103.11671","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:6876e38bc17aab549140a2fbcd43f857efd8e6bb19da4b7551f43dadd8075253","observation_id":"ad9a5a99-47c0-4895-9c72-3e0fbc123086","resolution":{"observed_at":"2026-08-03T17:28:41.597692Z","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":"2026-08-03T17:28:41.712116Z","title":"Visualizing data using t-sne.Journal of machine learning research, 9 (Nov):2579–2605, 2008","venue":null,"work_id":null,"year":2008},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.712116Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:ad95a551995839e986510a51869daea91e2d8dc2fab9af5a3e8c2c33ee248c98","observation_id":"9d9857a7-9fde-4c91-ba01-5238698ad562","resolution":{"observed_at":"2026-08-03T17:28:41.712116Z","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":"2026-08-03T17:28:41.842080Z","title":"Learning transferable visual models from natural language supervi- sion","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.842080Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:86c0d11749de3ebc41e4a34f09854bbbb1f7a884dbb8b9b4652c080d2f53d9d3","observation_id":"89a75560-120b-43e8-a898-f7c2dc284172","resolution":{"observed_at":"2026-08-03T17:28:41.842080Z","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":"2026-08-03T17:28:41.912138Z","title":"Towards to- tal recall in industrial anomaly detection","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.912138Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:044c2018880d271c4ce61e2c39f430bbf933573e2a8be4767af5eb5485fa10ec","observation_id":"dde70cc8-4fb5-48d3-a232-f5da0776f7bb","resolution":{"observed_at":"2026-08-03T17:28:41.912138Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2504.06740","last_updated":"2025-08-06T12:25:20Z","snapshot_observed_at":"2026-08-07T16:07:15.324840Z","submitted_at":"2025-04-09T09:52:04Z","title":"MultiADS: Defect-aware Supervision for Multi-type Anomaly Detection and Segmentation in Zero-Shot Learning","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2504.06740","snapshot_observed_at":"2026-08-03T17:28:41.985485Z","title":"Multiads: Defect-aware supervision for multi-type anomaly detection and segmentation in zero-shot learning.arXiv preprint arXiv:2504.06740, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:41.985485Z"},"links":{"cited_paper":"/paper/2504.06740","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:a90c79747933967d2916169c77649cf7e511c00194aa1711a86a57de6d66b695","observation_id":"e374e0fd-461a-484d-b874-4228f5a1ddcb","resolution":{"observed_at":"2026-08-03T17:28:41.985485Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2404.07983","last_updated":"2025-04-16T10:50:18Z","snapshot_observed_at":"2026-07-06T17:59:00.124173Z","submitted_at":"2024-04-11T17:58:06Z","title":"Two Effects, One Trigger: On the Modality Gap, Object Bias, and Information Imbalance in Contrastive Vision-Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2404.07983","snapshot_observed_at":"2026-08-03T17:28:42.092582Z","title":"Two effects, one trigger: on the modality gap, object bias, and information imbalance in contrastive vision-language representation learning.arXiv preprint arXiv:2404.07983, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.092582Z"},"links":{"cited_paper":"/paper/2404.07983","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:24c62d8ba6300304cc56143cce00deb6af020cd823be19523690645940f62e6d","observation_id":"3e104e80-3169-4f91-9d14-f6f87cd195d6","resolution":{"observed_at":"2026-08-03T17:28:42.092582Z","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":"2026-08-03T17:28:42.182275Z","title":"Attention guided anomaly localization in images","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.182275Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:92e4a5d797cc79849ca9fb11d4555b05ed96f40ff584fa481bdd66b14a9451be","observation_id":"0c006cc5-64fe-408b-9597-348e728a655d","resolution":{"observed_at":"2026-08-03T17:28:42.182275Z","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":"2026-08-03T17:28:42.349604Z","title":"Real-iad: A real-world multi-view dataset for benchmarking versatile industrial anomaly detec- tion","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.349604Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:1701bfa1a5bff663455859bed94b635e3f6d3e5b7e6e3c3df4da036f170cafb6","observation_id":"5cf7ff43-1741-4df2-8f7e-49f5c72b369b","resolution":{"observed_at":"2026-08-03T17:28:42.349604Z","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":"2026-08-03T17:28:42.467748Z","title":"Progressive vi- sual prompt learning with contrastive feature re-formation","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.467748Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:78b6eb08113cf2a9089175c2d90a1eed4f412da746db5ee7707f9c30ba247922","observation_id":"c16b8770-f821-4b02-84ee-9fdbc0db21a9","resolution":{"observed_at":"2026-08-03T17:28:42.467748Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2012.07122","last_updated":"2020-12-13T18:30:51Z","snapshot_observed_at":"2026-08-09T08:48:21.868623Z","submitted_at":"2020-12-13T18:30:51Z","title":"DFR: Deep Feature Reconstruction for Unsupervised Anomaly Segmentation","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2012.07122","snapshot_observed_at":"2026-08-03T17:28:42.570931Z","title":"Dfr: Deep feature re- construction for unsupervised anomaly segmentation.arXiv preprint arXiv:2012.07122, 2020","venue":null,"work_id":null,"year":2012},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.570931Z"},"links":{"cited_paper":"/paper/2012.07122","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:68a551e4fcbe003778b993d99f312e1100b61812879be8aa7973fb01516fc8fe","observation_id":"0de807ff-8669-422c-96ac-e003203eeea5","resolution":{"observed_at":"2026-08-03T17:28:42.570931Z","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":"2026-08-03T17:28:42.709999Z","title":"Long-horizon language-conditioned imita- tion learning for robotic manipulation.IEEE/ASME Trans- actions on Mechatronics, pages 1–12, 2025","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.709999Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:76b0575c6145e8ac61d17e65ded60ba081eec40356af70267f93a8fe5fd57267","observation_id":"ba0f64ce-c394-4ec3-9929-eeb7e69a9659","resolution":{"observed_at":"2026-08-03T17:28:42.709999Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2312.10807","last_updated":"2026-06-22T19:43:20Z","snapshot_observed_at":"2026-07-06T17:04:17.738921Z","submitted_at":"2023-12-17T20:13:20Z","title":"Bridging Language and Action: A Survey of Language-Conditioned Robot Manipulation","version":7},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2312.10807","snapshot_observed_at":"2026-08-03T17:28:42.820866Z","title":"Bridging language and action: A survey of language-conditioned robot manip- ulation.CoRR, abs/2312.10807, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.820866Z"},"links":{"cited_paper":"/paper/2312.10807","citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:4e99846933ce3aed79b357c1d3150819af149acbf8d9b49680a3d92f01e5636f","observation_id":"57fb63af-c805-4e8b-90db-b268c5caa787","resolution":{"observed_at":"2026-08-03T17:28:42.820866Z","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":"2026-08-03T17:28:42.918355Z","title":"Language-conditioned imitation learning with base skill pri- ors under unstructured data.IEEE Robotics and Automation Letters, 9(11):9805–9812, 2024","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:42.918355Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:47e1639fec56f3ba9f3e1b87db5f439500b250094047b4624ed58ba9eaa16c25","observation_id":"9b3dad24-c4e5-4e6f-914b-c6d4994791a9","resolution":{"observed_at":"2026-08-03T17:28:42.918355Z","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":"2026-08-03T17:28:43.050618Z","title":"Predicting the road ahead: A knowledge graph based foundation model for scene under- standing in autonomous driving","venue":null,"work_id":null,"year":2025},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.050618Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:4659e8d78ddb04a3b8ca02ba421f0074ba44a29e54748a4476bb91eb57a1f6af","observation_id":"4031bced-718c-4d9a-9656-b78a28ec2bec","resolution":{"observed_at":"2026-08-03T17:28:43.050618Z","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":"2026-08-03T17:28:43.207774Z","title":"Conditional prompt learning for vision-language mod- els","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.207774Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:9ab8b9ba3926dbcbce385422da30d0919b9b1f3c953a7f8401076c9d90f22acf","observation_id":"7312f5c2-52fb-4201-93a6-c0d5200a818a","resolution":{"observed_at":"2026-08-03T17:28:43.207774Z","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":"2026-08-03T17:28:43.356532Z","title":"Learning to prompt for vision-language models.In- ternational Journal of Computer Vision, 130(9):2337–2348,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.356532Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:7ad5f38e52242d8322949e5b259f446019190b2507a5414523e9234b961ade3b","observation_id":"eb0be11a-ad8c-4f57-962b-08507dafeaab","resolution":{"observed_at":"2026-08-03T17:28:43.356532Z","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":"2026-08-03T17:28:43.459493Z","title":"Pad: A dataset and benchmark for pose-agnostic anomaly detection.Advances in Neural Information Processing Systems, 36:44558–44571,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.459493Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:83b3d10c793e8cd35f93805c58f8543192b384b1fa006efaa231221d18af4b81","observation_id":"e0204525-47c7-46ed-8047-57998c912043","resolution":{"observed_at":"2026-08-03T17:28:43.459493Z","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":"2026-08-03T17:28:43.584073Z","title":"Anomalyclip: Object-agnostic prompt learn- ing for zero-shot anomaly detection.arXiv preprint arXiv:2310.18961, 2023","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.584073Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:6fbf2757e28fc0c9716468f07641684e09e88056a6b208900989521d7d3415e5","observation_id":"6dcd9ed7-2d6f-4d19-93b7-dfeffa44e1c5","resolution":{"observed_at":"2026-08-03T17:28:43.584073Z","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":"2026-08-03T17:28:43.692925Z","title":null,"venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.692925Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:00309da10c162d5c4e4a310478e229607df679d4a2437e5c2c41cbe477a8967d","observation_id":"ed3812d7-7ab3-4342-b584-fd6516be398f","resolution":{"observed_at":"2026-08-03T17:28:43.692925Z","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":"2026-08-03T17:28:43.822761Z","title":"Fine-grained abnormality prompt learning for zero- shot anomaly detection.arXiv preprint arXiv:2410.10289,","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.822761Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:3bb02b7cd9cf6d98f2c92450809bb83e77d04929aced7332acc2d618a0bc1f97","observation_id":"8ae772ad-7cce-48b0-a90b-15169493f011","resolution":{"observed_at":"2026-08-03T17:28:43.822761Z","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":"2026-08-03T17:28:43.921994Z","title":"Unsupervised anomaly detection with an enhanced teacher for student- teacher feature pyramid matching","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:43.921994Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:1bf03f73a1391a0d4fa487f976e3b5e6759e2a4466b396ff89c099c99acf02fd","observation_id":"389300ba-d8a8-4113-9446-47d74956019f","resolution":{"observed_at":"2026-08-03T17:28:43.921994Z","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":"2026-08-03T17:28:44.041860Z","title":"Spot-the-difference self-supervised pre- training for anomaly detection and segmentation","venue":null,"work_id":null,"year":2022},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.041860Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:3f8c35e7f43b2e5aaa68b9547794c2aecbd620bcba076bb368b9f9ac2959ee7f","observation_id":"be436fc3-e23e-48b7-b7f8-1b872e6d2d12","resolution":{"observed_at":"2026-08-03T17:28:44.041860Z","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":"2026-08-03T17:28:44.177903Z","title":"Table 5 summarizes key statistics of these datasets, includ- ing the number of distinct product classes and the distri- bution of normal and anomalous samples","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.177903Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:65b601f83110cfb2b8b7e397eddddabc04b8974b1f1c8d3d51ca422aeae54a9a","observation_id":"5cf086cf-59fb-4111-9827-7d7b6ed85d4c","resolution":{"observed_at":"2026-08-03T17:28:44.177903Z","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":"2026-08-03T17:28:44.303097Z","title":"The results of the baselines are taken directly from the respective papers","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.303097Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:f527824db7f986cfe0ff2a7f74e29fdc62ce7e3bdacea446a2f18373b5faed57","observation_id":"0ed12132-f189-47b3-bd26-79ddde6c37ae","resolution":{"observed_at":"2026-08-03T17:28:44.303097Z","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":"2026-08-03T17:28:44.437875Z","title":"implementation details","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.437875Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:5cdb93f056fdca18b788293ec07000b26d0be06355776d3ab839ea5681d12527","observation_id":"479158d3-072a-448d-bb93-08495c26f4b0","resolution":{"observed_at":"2026-08-03T17:28:44.437875Z","resolver_source":null,"status":"malformed_identifier"},"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":"2026-08-03T17:28:44.602283Z","title":"contam- ination","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.602283Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:4964ec993954e03d7605b6020811691bbc575ed6e2ce40b4f153359018da6769","observation_id":"f58d27b2-ad7f-4906-a445-22c9c89a52cd","resolution":{"observed_at":"2026-08-03T17:28:44.602283Z","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":"2026-08-03T17:28:44.745087Z","title":"We present six examples of object from VisA, MPDD, MVTec-AD, and Real-IAD datasets","venue":null,"work_id":null,"year":null},"citing_paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation","version":3},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-03T17:28:44.745087Z"},"links":{"citing_paper":"/paper/2512.09446"},"observation_digest":"sha256:d4ec929ac0a8923ae78c4a756756892b4023ec9d6635e5a6622db6726e046a59","observation_id":"04d77d71-07c3-416e-997a-095ab415f5df","resolution":{"observed_at":"2026-08-03T17:28:44.745087Z","resolver_source":null,"status":"malformed_identifier"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}}],"paper":{"arxiv_id":"2512.09446","last_updated":"2026-07-02T13:02:35Z","latest_version":3,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-07T10:09:47.142992Z","submitted_at":"2025-12-10T09:19:17Z","title":"Defect-aware Hybrid Prompt Optimization via Progressive Tuning for Zero-Shot Multi-type Anomaly Detection and Segmentation"},"reference_resolution":{"displayed":50,"state_counts":{"malformed_identifier":2,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":48,"verified_exact":0,"verified_fuzzy":0},"total_outbound_references":50},"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":"2026-08-09T06:31:02.800959+00:00","source":"crossref"},{"observed_at":"2026-08-09T06:30:57.326959+00:00","source":"retraction_watch"}],"thesis":"As of 9 August 2026, this Paper Citation Record lists 50 of 50 outbound references and 1 inbound Pith citation observation for arXiv:2512.09446."}