{"as_of":"2026-08-09T21:43:00Z","caps":{"database_statements":6,"inbound":100,"outbound":100},"context_digest":"sha256:992d56c83cde5b375e30f88b7b53c128373cff5863ee8a040961316aee60bb2f","coverage":[{"denominator":72,"lane":"reference_resolution","note":"Typed states for the displayed outbound observations.","records_observed":72,"source":"paper_references, paper_reference_links","source_observed_at":"2026-08-05T14:01:41.595916Z","state":"measured"},{"denominator":72,"lane":"standing_notices","note":"One-hop event checks from named stored sources.","records_observed":72,"source":"scholarly_work_events, retraction_status_cache","source_observed_at":"2026-08-09T06:31:02.800959+00:00","state":"measured"},{"denominator":0,"lane":"inbound_itemization","note":"Pith citing papers itemized under the disclosed page cap.","records_observed":0,"source":"paper_references, paper_reference_links","source_observed_at":null,"state":"measured"},{"denominator":1,"lane":"external_citation_measurements","note":"A source-named dated measurement, never combined with another source.","records_observed":0,"source":"cited_works","source_observed_at":null,"state":"measured"}],"external_citation_measurements":[],"inbound":[],"links":{"evidence":"/evidence","html":"/paper/2508.21795/citation-record","integrity":"/paper/2508.21795/integrity","json":"/paper/2508.21795/citation-record.json","paper":"/paper/2508.21795"},"outbound":[{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:46.044884Z","title":"Deep anomaly detection using geometric transformations,","venue":null,"work_id":"6f44065e-3e51-43b0-ad73-958285eea61c","year":2018},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":1,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.002798Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1ed38165dd0facbd99085a9a25816346a74924865aa2235b1376f9969fc46085","observation_id":"e1fe2e89-1e69-4df3-bb28-c7c29e9e8d5c","resolution":{"observed_at":"2026-08-05T14:01:46.048265Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:46.034731Z","title":"Towards total recall in industrial anomaly detection,","venue":null,"work_id":"48e941b9-c111-446c-9478-1d41bdd02058","year":2022},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":2,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.068003Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:9df270f550683b4300f33cd92462dcf9304c85092bb8717f5ff7fb92b28a9561","observation_id":"711a5257-e792-4665-b5fd-7e4a82daae79","resolution":{"observed_at":"2026-08-05T14:01:46.038511Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:46.024175Z","title":"Promptad: Learning prompts with only normal samples for few-shot anomaly detection,","venue":null,"work_id":"cb5d372e-01a5-4223-800a-7a6d330cba9f","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":3,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.137213Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:773252f9eb2336c9db081a940e66b6b65c68e483a5e8c73ebb44f5c7bcd4ebfa","observation_id":"e0361a9a-c0fa-4892-afb2-402c6b8e00ba","resolution":{"observed_at":"2026-08-05T14:01:46.028300Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:46.013629Z","title":"Mvtec ad–a comprehensive real-world dataset for unsupervised anomaly detection,","venue":null,"work_id":"062bded8-c289-4450-8c54-7eca32230f25","year":2019},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":4,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.328970Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:c6ed1231e58f2e24400687095d964ec7bee4f7c7a495a0320a01943f75ebcd95","observation_id":"11635f1c-d081-4871-8b8b-21a072880aca","resolution":{"observed_at":"2026-08-05T14:01:46.017754Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:46.002253Z","title":"Spot-the- difference self-supervised pre-training for anomaly detection and seg- mentation,","venue":null,"work_id":"ec87037d-8c34-44b7-afea-2a3633e7f863","year":2022},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":5,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.400347Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:d2fd370f5fc148f6d703ff061dd9255f6f7a6384e8ddc714ddea3b4ef58ae27d","observation_id":"29529d45-4a9e-482c-b26c-cb68da92e2f2","resolution":{"observed_at":"2026-08-05T14:01:46.005851Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.991730Z","title":"Be- yond dents and scratches: Logical constraints in unsupervised anomaly detection and localization,","venue":null,"work_id":"86822e3a-3567-4817-960d-ceeaf9205727","year":2022},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":6,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.531784Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:db1c6f5531cc706a2f84d27f59cd19ab2f414ce6ce9e96c8dbd5830dfc9dc5b2","observation_id":"d08fe6c0-65df-4cfb-988f-f3d469fc6e8f","resolution":{"observed_at":"2026-08-05T14:01:45.995315Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.980155Z","title":"Correcting deviations from normality: A reformulated diffusion model for multi- class unsupervised anomaly detection,","venue":null,"work_id":"bcabd182-4548-44db-8a49-9b82c219a437","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":7,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.597734Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:791a42d416feac109359b10c2396725d334474e82aec3099754dbd06334bf16d","observation_id":"82977ecf-1782-44ab-bc28-63653719dae4","resolution":{"observed_at":"2026-08-05T14:01:45.984749Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.969321Z","title":"A cognitive memory-augmented network for visual anomaly detection.,","venue":null,"work_id":"0db03708-628f-4dd9-95b9-6ad6a5000772","year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":8,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.688191Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:992490afb0e5848a28d2352840f7e78b5b3a14d76e0c19fb5846e45b1283b2e1","observation_id":"9240be01-eafe-43eb-9244-e0876be237f4","resolution":{"observed_at":"2026-08-05T14:01:45.972905Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.959094Z","title":"Divide- and-assemble: Learning block-wise memory for unsupervised anomaly detection,","venue":null,"work_id":"ada15465-0e19-4a4d-811e-2931b1bb384d","year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":9,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.786606Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:7d61a23071a9dd37277660fa5ce9eb9f79b9d965cadca0eb06677d46aca40364","observation_id":"e521775f-218b-4417-a53c-be7916b4b6dd","resolution":{"observed_at":"2026-08-05T14:01:45.962699Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.949342Z","title":"Rethinking autoencoders for medical anomaly detection from a theoretical perspective,","venue":null,"work_id":"5c440716-620a-4bae-9a14-d1fa506ce445","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":10,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:35.998654Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:826de1d180865dbc5dbbcb0c67fe22d538cefde9d8f11c4da8d95c2148021ec2","observation_id":"b97690d9-81a8-44e1-838c-edc0ac356493","resolution":{"observed_at":"2026-08-05T14:01:45.952568Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.939201Z","title":"Realnet: A feature selection network with realistic synthetic anomaly for anomaly detection,","venue":null,"work_id":"3d30eda2-1e42-4341-a137-a540e659e62b","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":11,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.155396Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:cfb014eb1b0bb3c713b1f7a2c2c12592ed2a5734b223725b1fdd38811c3c6f5e","observation_id":"dab56856-6c44-427e-bf1b-9e4849dfa81e","resolution":{"observed_at":"2026-08-05T14:01:45.942789Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.929276Z","title":"A diffusion-based framework for multi-class anomaly detection,","venue":null,"work_id":"e9ef94c1-2e70-4f43-bdf5-debf86ef4b4a","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":12,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.291048Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:dcd3cee3ccce69bc4750fe3e15f5a9f0f7d014dcbf1a97479f87624df31d7ec6","observation_id":"815d9be0-75d2-4291-b9dc-f8d1aa5db637","resolution":{"observed_at":"2026-08-05T14:01:45.932831Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.918887Z","title":"Residual denoising diffusion models,","venue":null,"work_id":"758ea2c1-e996-4410-b2a0-4046da427b30","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":13,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.439395Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:12c5192852fc3ddf705585efef4af53faf0b6689f924c7fb3260171349fd0c01","observation_id":"d0d40bd1-393d-45db-b274-a7b0021c0e17","resolution":{"observed_at":"2026-08-05T14:01:45.923098Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.907919Z","title":"Glad: Towards better reconstruction with global and local adaptive diffusion models for unsupervised anomaly detection,","venue":null,"work_id":"5cc61daa-6fc1-4cd1-a16b-7ac05f65e9dc","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":14,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.541933Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:49dc548fa0c1da4d6cf4c25b93297f68487d4fb190a571056866d8e1e4d3e53c","observation_id":"3d3593f4-08ab-46db-9e80-dc1e817cdf43","resolution":{"observed_at":"2026-08-05T14:01:45.912052Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.896283Z","title":"Anomaly detection via reverse distillation from one-class embedding,","venue":null,"work_id":"facc7d69-4660-4388-be8c-54ae5f079f75","year":2022},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":15,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.646057Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:9e19ae89fd73ccda1490d04a290bb781b57363d5cdbb4b055445298207f46797","observation_id":"7729c6e1-f0dd-445c-b1fe-9f9cf79673b3","resolution":{"observed_at":"2026-08-05T14:01:45.900400Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.884619Z","title":"Dual-modeling decouple distillation for unsupervised anomaly detection,","venue":null,"work_id":"cda06ede-3f8d-46b2-9d5f-7060badefbc0","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":16,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.759335Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:550d50fd4f868f72a9c12ff76a82f9d5df151d080a970c03d8f3a764dfe732da","observation_id":"9aebd1ca-09a3-4c91-a2e5-b4a4d0eb0e64","resolution":{"observed_at":"2026-08-05T14:01:45.888954Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.874687Z","title":"Feature-constrained and attention-conditioned distillation learning for visual anomaly detection,","venue":null,"work_id":"1188fd31-7e85-4100-831a-27d23e041eef","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":17,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.854532Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:b5fc22a11082a9cbe2f4b970db8ae8cc862ae6e4526de3aeb24a8c11de960206","observation_id":"40470317-f0de-429d-a704-5f50c136b934","resolution":{"observed_at":"2026-08-05T14:01:45.878458Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.864984Z","title":"Aekd: Unsupervised auto- encoder knowledge distillation for industrial anomaly detection,","venue":null,"work_id":"ab10d495-dee7-4e32-9248-ad9d5324dad1","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":18,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:36.936905Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:0b56c592cfd585938c871aa86827cf6388fa48641d0af9d3d0c9c15ee1776b78","observation_id":"10ffed13-a929-4b74-a59e-e1020781110e","resolution":{"observed_at":"2026-08-05T14:01:45.868290Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.854998Z","title":"Pushing the limits of fewshot anomaly detection in industry vision: Graphcore,","venue":null,"work_id":"190b7368-9690-4a8c-9068-fb2b9de8f0b7","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":19,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.011211Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:3d1a28aee1d8e429e2bf1b5a558cc0b6aafcac8cd9afe3ec6aafbbc353221da3","observation_id":"706bc590-2f1c-4e38-9574-f32083b97895","resolution":{"observed_at":"2026-08-05T14:01:45.859042Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.844957Z","title":"Pni: industrial anomaly detection using position and neighborhood information,","venue":null,"work_id":"7f88750c-4bcb-4700-8e09-90f9531d1e9d","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":20,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.084555Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:caf5bcf212b61edb1613c16a952b0605add1f5522867ecb7333c9fdd7f894a4e","observation_id":"eb3fab2b-e7b1-4456-9a8f-f6343f4d7dbc","resolution":{"observed_at":"2026-08-05T14:01:45.848512Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.835312Z","title":"Progressive bound- ary guided anomaly synthesis for industrial anomaly detection,","venue":null,"work_id":"956ac3d7-c415-42b7-b1ef-49e987e91890","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":21,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.202838Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:eb0199675af09ce25be034201a36dc0789357e33e95c9a01665e33939eec0a95","observation_id":"a6a55bf0-49ab-4b15-96ce-e720ee3425c9","resolution":{"observed_at":"2026-08-05T14:01:45.838479Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.824402Z","title":"Inter-realization channels: Unsupervised anomaly detection beyond one-class classification,","venue":null,"work_id":"d1553d7b-e6a9-4f33-85d8-ec36a54171a1","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":22,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.305926Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:85512ea652e4f88e3d22bec52b397c285c6a07591ccdf857ca7193717d8d3950","observation_id":"2dea5e79-c2a0-4d39-a679-219346543825","resolution":{"observed_at":"2026-08-05T14:01:45.828005Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.812729Z","title":"Reconpatch: Contrastive patch representation learning for industrial anomaly detection,","venue":null,"work_id":"96ae9e3a-46da-4566-b4d0-e21f17d2cb5c","year":2052},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":23,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.380934Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:b44e0d65fb4d1b515c3ef85cce452fef3409aae12e6668c1c4b8ae1b9eda7e34","observation_id":"ab44e019-144e-4853-bb96-a82c880e0c53","resolution":{"observed_at":"2026-08-05T14:01:45.817423Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.801920Z","title":"A reconstruction-based feature adaptation for anomaly detection with self-supervised multi-scale aggregation,","venue":null,"work_id":"dc2e8805-12b2-43a5-bc71-5c404177f07e","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":24,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.470097Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:43bf9831b8618b2d88489303702251126772d63aab8f6bf1d31ca16696fee5b6","observation_id":"ada09d41-7e72-498f-96b8-8e36fea2360d","resolution":{"observed_at":"2026-08-05T14:01:45.806034Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.791305Z","title":"Towards training-free anomaly detection with vision and language foundation models,","venue":null,"work_id":"356ecd3b-9601-480a-80c1-dd805a928bef","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":25,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.546581Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:e977beaec4587c9fdec6eb68185286feeea3c7838fe68e78638d74571c07f1fb","observation_id":"d6358c83-76ad-43e9-b016-37d0380bf1ab","resolution":{"observed_at":"2026-08-05T14:01:45.795502Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.780027Z","title":"Space: Spatial- aware consistency regularization for anomaly detection in industrial applications,","venue":null,"work_id":"8b048b06-c3f0-4640-80bc-9a5681cdd5cc","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":26,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.642164Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:033ef53432243f37c5b2a9a1940921d171d4b26ed0ef0b78fbf653d3e8aaef6b","observation_id":"7bbdb976-36d9-4215-be70-119cad7b3495","resolution":{"observed_at":"2026-08-05T14:01:45.783684Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.768228Z","title":"Contextual affinity distillation for image anomaly detection,","venue":null,"work_id":"c0db296c-427d-40df-b510-0f9d39d2e812","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":27,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.773313Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:4ed5b79cafffa5ba278485a138bcfe27e43877e0da4a381bb80fe71db0801d92","observation_id":"33c0f4ae-3cbb-4df8-9637-254530400e05","resolution":{"observed_at":"2026-08-05T14:01:45.772392Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.758560Z","title":"Efficientad: Accurate visual anomaly detection at millisecond-level latencies,","venue":null,"work_id":"c7b8afa5-256e-4469-a7fc-97d09d8a792b","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":28,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.922670Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:6df6ec3ec7a286e718ff942c0749190ff0e8a68e4def504da023bc7fb8518a3e","observation_id":"9407b804-8367-4aec-8fad-c6b69b25ff71","resolution":{"observed_at":"2026-08-05T14:01:45.762076Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.747784Z","title":"Few shot part segmentation reveals compositional logic for industrial anomaly detection,","venue":null,"work_id":"1a836a73-aabe-4837-8910-db331f35a6e7","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":29,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:37.996942Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:154074df970cd019151bb0fd3865c2c1a01d1d9100e41ed8890f80ad2b5148d1","observation_id":"6feed949-31bb-4daf-a9d3-8705dd281b65","resolution":{"observed_at":"2026-08-05T14:01:45.751452Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.737503Z","title":"Univad: A training-free unified model for few-shot visual anomaly detection,","venue":null,"work_id":"2de0befd-f161-4804-898f-6917085e35c5","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":30,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.079274Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:79976aaabb779f23c6877bb1bbc30c022e75574a03e0dee02a8bd8f511af7872","observation_id":"00fafb0d-dc01-44c9-bf09-eae151d2bac5","resolution":{"observed_at":"2026-08-05T14:01:45.741130Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.725067Z","title":"Sam- lad: Segment anything model meets zero-shot logic anomaly detection,","venue":null,"work_id":"f47216af-0bbf-40c5-832d-b893ccf1a59f","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":31,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.179037Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:ef74a2e2fbe1836c8c262ef80df4bf398c6baf2f3f2a23b5b1cf50bd24c0312e","observation_id":"7a3776d8-3078-4e99-adfc-4a51d159c4af","resolution":{"observed_at":"2026-08-05T14:01:45.730085Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.714241Z","title":"Visual anomaly detection via partition memory bank module and error estimation,","venue":null,"work_id":"5961fb04-fbe8-40e9-98ea-20ac7e8c6e2e","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":32,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.284957Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:afa90e19c96d8bebf2b32321798cc3c416cec3af4598eca2a1df66d9c489c7f8","observation_id":"c76563b5-e380-4182-8de6-f067500cb352","resolution":{"observed_at":"2026-08-05T14:01:45.717723Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.703229Z","title":"Outlier-probability-based feature adaptation for robust unsupervised anomaly detection on contaminated training data,","venue":null,"work_id":"d9a49bd3-c3f0-4a8a-91ed-165d500ea845","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":33,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.405692Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:bbea9867c4c3aa31a422f9a98a9ceb31c20a6b8ca653ce4eac7e7cb1e9c3d6ae","observation_id":"396ed6cf-916c-427b-a6e3-62dc5a101236","resolution":{"observed_at":"2026-08-05T14:01:45.707617Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.692994Z","title":"Gaussian mixture models,","venue":null,"work_id":"7b901753-f67c-4c8f-b6cf-e661c75c79c9","year":2015},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":34,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.520434Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:6f2a8e4ee5793fad25749161a6afe9fec1ac95cd4a74352c7ec9ae3af54abd09","observation_id":"9b7c8f99-b728-4b23-9985-63eea641fba2","resolution":{"observed_at":"2026-08-05T14:01:45.696464Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.683266Z","title":"Anomalygpt: Detecting industrial anomalies using large vision-language models,","venue":null,"work_id":"9ec26d7c-05b4-4b5e-8b29-c451c105e563","year":1932},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":35,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.629882Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:5b0bde1987f21ed3bd6c3fab2d55ee0e2713b8736d427d85404be387834f88af","observation_id":"75397760-5222-464c-b851-ffed40efc653","resolution":{"observed_at":"2026-08-05T14:01:45.686723Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.672282Z","title":"Focusclip: Focusing on anomaly regions by visual-text discrepancies,","venue":null,"work_id":"6ebb7e51-cc74-4271-885d-fe2f11bbdf65","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":36,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.747593Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:b82d60ab77c04239f23c9c7e21ac1782cc4c64c5e4388b0249b4ea82e37af0e9","observation_id":"5b51e9dc-3ad4-4f50-a293-69e00966813b","resolution":{"observed_at":"2026-08-05T14:01:45.675961Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.662689Z","title":"Crepe: Can vision-language foundation models reason compositionally?,","venue":null,"work_id":"45d54675-3915-4cfa-9823-65a70397b203","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":37,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.874035Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:f89f38f2d5ded49b0632ae2e5e02bd0cbca2370b5b4b8a782747f8387a9090ec","observation_id":"02a559a1-d0a9-4e67-90e7-c06a1fbaf7fa","resolution":{"observed_at":"2026-08-05T14:01:45.666242Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.653754Z","title":"Aa-clip: Enhancing zero-shot anomaly detection via anomaly-aware clip,","venue":null,"work_id":"e7096ba2-6a50-4dc4-a424-15657f57f09b","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":38,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:38.976391Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:645b171535583460bf5914a399ecfe2108fedc3161fcfa16e8214a6fa3ffa65c","observation_id":"b2fc0584-8aeb-4243-9e45-0f3765b79a0a","resolution":{"observed_at":"2026-08-05T14:01:45.656732Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.644504Z","title":"Compositional chain- of-thought prompting for large multimodal models,","venue":null,"work_id":"854fcdec-e304-42d8-925c-bf210ec4f782","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":39,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.061472Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:8d2c3774f42ce7d1290405eb3b5b9efa87c39216ebdf6e7129c5b80d2fa14219","observation_id":"76c86c18-f420-429a-9c76-fd0f450efb9c","resolution":{"observed_at":"2026-08-05T14:01:45.647917Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.635660Z","title":"Logicqa: Logical anomaly detection with vision language model generated questions,","venue":null,"work_id":"982e4106-cee0-4bff-a016-c773ce5a6b32","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":40,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.157336Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:7f76ab391cf5947f44032f953f6971563321010cf8a3d3b7b2e3383d1403adf1","observation_id":"e6b4ff05-b977-4f1a-9545-f2e4e0d9e554","resolution":{"observed_at":"2026-08-05T14:01:45.638559Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.484648Z","title":"Inves- tigating compositional challenges in vision-language models for visual grounding,","venue":null,"work_id":"ab94a2cb-ddfb-4cd9-aff9-b1dda11e514e","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":41,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.253158Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:4b1ebb1c814a86770ad91b7920328876fff4056da5e49aa52e1d99e85a548fc6","observation_id":"5636002e-f810-4887-a027-dd21f51c7885","resolution":{"observed_at":"2026-08-05T14:01:45.617505Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.281889Z","title":"Segment anything,","venue":null,"work_id":"2d444c5b-ee5b-491f-a081-80b216452818","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":42,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.349078Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:d0ca39bb526b4fbbf9f5db19226cfba79ca7fd82f138484c29aa9a27072860ca","observation_id":"e8d4b6b6-ebc3-4d02-bc43-af2dfefaa10a","resolution":{"observed_at":"2026-08-05T14:01:45.413613Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:45.038074Z","title":"Learning transferable visual models from natural language supervision,","venue":null,"work_id":"60cf4cc5-bcdb-4808-9854-4b7d7ae21f7c","year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":43,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.431761Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:f25571b436a5bb9d770a56f94f1d409fd1f8fc3865f10d1dcf60bc4bbbea8e45","observation_id":"067a52a1-b60f-4da4-86ff-d82bc1039fe2","resolution":{"observed_at":"2026-08-05T14:01:45.114146Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":{"arxiv_id":"2409.12191","last_updated":"2024-10-03T15:54:49Z","snapshot_observed_at":"2026-08-06T05:35:29.109022Z","submitted_at":"2024-09-18T17:59:32Z","title":"Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2409.12191","snapshot_observed_at":"2026-08-05T14:01:39.539111Z","title":"Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":44,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.539111Z"},"links":{"cited_paper":"/paper/2409.12191","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:b521697b17095a7fa6a2d016039ccf7bee57952f6cbe9485667e5d11c6696581","observation_id":"fdf19a75-9c50-4cbd-95d3-31f1ff1fec0f","resolution":{"observed_at":"2026-08-05T14:01:39.539111Z","resolver_source":null,"status":"unresolved"},"standing_notice":{"events":[],"reason":"canonical_work_link_unavailable","source_receipts":[],"state":"unavailable"}},{"citation":{"cited_paper":{"arxiv_id":"2401.14159","last_updated":"2024-01-25T13:12:09Z","snapshot_observed_at":"2026-07-06T17:20:25.138890Z","submitted_at":"2024-01-25T13:12:09Z","title":"Grounded SAM: Assembling Open-World Models for Diverse Visual Tasks","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2401.14159","snapshot_observed_at":"2026-08-05T14:01:39.654341Z","title":"Grounded sam: Assembling open-world models for diverse visual tasks,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":45,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.654341Z"},"links":{"cited_paper":"/paper/2401.14159","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:807fff8a1a11f673ce4cf158899e40bbb29be60959b18a4ba01e991330c9c4a6","observation_id":"a68e2ee9-8dad-4a2c-ba88-e41773b8fedb","resolution":{"observed_at":"2026-08-05T14:01:39.654341Z","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":"2026-08-05T14:01:44.761851Z","title":"Towards zero- shot anomaly detection and reasoning with multimodal large language models,","venue":null,"work_id":"0db8cc85-58bb-4f13-a37d-528446683f8b","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":46,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.767666Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:41e0ef579da99ec57d08534be2ca93e9473c17da8141a27d124e83de417f9cac","observation_id":"db2cd765-822b-4775-95e7-eacd48b9af95","resolution":{"observed_at":"2026-08-05T14:01:44.892636Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":{"arxiv_id":"2410.21276","last_updated":"2024-10-25T17:43:01Z","snapshot_observed_at":"2026-07-06T02:11:23.670680Z","submitted_at":"2024-10-25T17:43:01Z","title":"GPT-4o System Card","version":1},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2410.21276","snapshot_observed_at":"2026-08-05T14:01:39.867318Z","title":"Gpt-4o system card,","venue":null,"work_id":null,"year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":47,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.867318Z"},"links":{"cited_paper":"/paper/2410.21276","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:fd8b3c80ce58af599de6ccb73bd1af114e74a1cea0e670f5af9d8d41ab0e5d45","observation_id":"555f6815-e6c5-43c3-9720-cf2218c75d22","resolution":{"observed_at":"2026-08-05T14:01:39.867318Z","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":"2026-08-05T14:01:44.614771Z","title":"Mitigating hallucination in large multi-modal models via robust instruction tuning,","venue":null,"work_id":"9a509578-ba8f-4b17-93d2-52ac146e198c","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":48,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:39.914922Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:903c9e19dc3f8f607540f8c8a996666cfa87da4531dce796e41aa7dda5a14fc1","observation_id":"5ff0b8a0-6f68-4cc9-8b88-60bbb231d64a","resolution":{"observed_at":"2026-08-05T14:01:44.710663Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:44.517242Z","title":"Dinov2: Learning robust visual features without supervision,","venue":null,"work_id":"8c0fe9e1-198a-4503-abc4-fdc4fe0f1618","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":49,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.028881Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1f757fdf56ceeaaf80c8c55791610f5c01baf2d0a2ca9d4530a95f3b157500c7","observation_id":"33de0a48-1291-492a-9dc9-a3bad3605ee6","resolution":{"observed_at":"2026-08-05T14:01:44.560367Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":{"arxiv_id":"2310.08825","last_updated":"2024-03-08T02:49:12Z","snapshot_observed_at":"2026-08-07T23:37:32.488201Z","submitted_at":"2023-10-13T02:41:55Z","title":"From CLIP to DINO: Visual Encoders Shout in Multi-modal Large Language Models","version":3},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2310.08825","snapshot_observed_at":"2026-08-05T14:01:40.129707Z","title":"From clip to dino: Visual encoders shout in multi-modal large language models,","venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":50,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.129707Z"},"links":{"cited_paper":"/paper/2310.08825","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:8620e40dae677cce6fe78e3b0ebf6c7d36a487f0401988f1a6a3e7561b87ba54","observation_id":"a92a6dea-f979-4319-b35a-d63074e2910a","resolution":{"observed_at":"2026-08-05T14:01:40.129707Z","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":"2026-08-05T14:01:44.329537Z","title":"Deep learning-based defect detection of metal parts: evaluating current meth- ods in complex conditions,","venue":null,"work_id":"ff7c4323-8f2c-40b7-8295-2eafd03a02ae","year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":51,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.250841Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:a021db33f8f6129cc06c160a1712d86c4d911aeadefb855071824ade975cc768","observation_id":"a0597b7d-99b5-4678-960c-5af1e9a9d0f4","resolution":{"observed_at":"2026-08-05T14:01:44.423332Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":{"arxiv_id":"2107.02314","last_updated":"2021-09-12T20:26:52Z","snapshot_observed_at":"2026-08-06T01:56:28.054412Z","submitted_at":"2021-07-05T23:12:06Z","title":"The RSNA-ASNR-MICCAI BraTS 2021 Benchmark on Brain Tumor Segmentation and Radiogenomic Classification","version":2},"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":null,"pith_arxiv_id":"2107.02314","snapshot_observed_at":"2026-08-05T14:01:40.328658Z","title":"The rsna-asnr-miccai brats 2021 benchmark on brain tumor segmentation and radiogenomic classification,","venue":null,"work_id":null,"year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":52,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.328658Z"},"links":{"cited_paper":"/paper/2107.02314","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:69f0f1998a256b73a1d86323deb8c9d1d001d26349a7504573065346e1c914e6","observation_id":"a79ee202-2721-42a2-b166-1dcf528008fe","resolution":{"observed_at":"2026-08-05T14:01:40.328658Z","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":"2026-08-05T14:01:44.152330Z","title":"Miccai multi-atlas labeling beyond the cranial vault–workshop and challenge,","venue":null,"work_id":"0fb5268a-4d63-47c6-a8b0-cafcae366511","year":2015},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":53,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.389727Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:479c851afd07b007ad3c8520196f2857095e95d34c040d81c5b5619649ff8676","observation_id":"307003f2-c282-42c7-a087-ccbf053cc75b","resolution":{"observed_at":"2026-08-05T14:01:44.236849Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:44.024444Z","title":"Automated segmentation of macular edema in oct using deep neural networks,","venue":null,"work_id":"6d68ba9f-7ae7-4741-b8a2-64d66e43bd1b","year":2019},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":54,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.448432Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:b4ffff8ccf6515f8de41c9227fb5e5b44ac6c8d7cfcac5bdf48ef52c8ac6d970","observation_id":"540b938e-f256-477e-92f8-94e1f080e9c6","resolution":{"observed_at":"2026-08-05T14:01:44.088247Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.830554Z","title":"Gestalt pattern matching","venue":null,"work_id":"890b854a-677d-4b1c-a0b1-c4d7d0990fc8","year":1983},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":55,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.495403Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1b5cd0dc00a27fefad22ffd1ee72787af115a5d62a8caee285131f6220d3370c","observation_id":"0f0f56ff-ee54-42e0-b8b1-1c8d632627a6","resolution":{"observed_at":"2026-08-05T14:01:43.908867Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.700570Z","title":"Logical: Towards logical anomaly synthesis for unsupervised anomaly localization,","venue":null,"work_id":"89b835f2-a321-41ba-be12-ecd98a033163","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":56,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.545596Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1cf18452f3ae0278a90af8183265ea1792e623b85a4ab76eab8a92ee2bde6c2e","observation_id":"525725f3-b076-4fb7-be6b-592bffc41c32","resolution":{"observed_at":"2026-08-05T14:01:43.768418Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.583096Z","title":"Draem-a discriminatively trained reconstruction embedding for surface anomaly detection,","venue":null,"work_id":"2750f185-aee8-46d0-9eef-b5e80c682a65","year":2021},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":57,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.626482Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:7c2eb7b4d0fb927892e5843e4cd1cd5d11c0953686524dfafb9d3c3d01c5ea0e","observation_id":"91bb5ee8-05de-4f44-b680-38860d7c3285","resolution":{"observed_at":"2026-08-05T14:01:43.647309Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.458678Z","title":"Omnial: A unified cnn framework for unsupervised anomaly localization,","venue":null,"work_id":"512bdfdf-8b80-4e11-90ef-fddba8ddad8a","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":58,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.691720Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1d6a793ba362b9c5ee121115f5d74882398485dc3ffe0b86c70113d99f2feb0c","observation_id":"082b8e9b-3292-417a-8e3f-ddd1c023468c","resolution":{"observed_at":"2026-08-05T14:01:43.511866Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.287520Z","title":"Generalad: Anomaly detection across domains by attending to distorted features,","venue":null,"work_id":"692e43b9-96d9-4714-8de1-11148f0e1101","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":59,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.713011Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:981644afce1190f17e5375c569cdfd4cf0bf3f7516972ae9ff06f4fbc0cbb327","observation_id":"62045f19-d5e6-472a-8a84-a55d31972038","resolution":{"observed_at":"2026-08-05T14:01:43.366715Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:43.112227Z","title":"Winclip: Zero-/few-shot anomaly classification and segmentation,","venue":null,"work_id":"b83831dc-363f-4c86-9fbf-28d6ee72f058","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":60,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.771442Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:ab7e2d89ab46a55e1f4796635a1cef9c76b1e43bf13f39692f8c7234e3eccba2","observation_id":"18f3dc6b-491d-4e64-848c-99f05ffbf632","resolution":{"observed_at":"2026-08-05T14:01:43.199432Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"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":"2026-08-05T14:01:40.849367Z","title":null,"venue":null,"work_id":null,"year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":61,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.849367Z"},"links":{"cited_paper":"/paper/2305.17382","citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:583dbedbfd6ade0a05fa3ef4dd3a843138ffd257093a3539b510d7ec66ad6587","observation_id":"70198286-67da-4ce5-a02a-93fc25977503","resolution":{"observed_at":"2026-08-05T14:01:40.849367Z","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":"2026-08-05T14:01:42.940172Z","title":"Anomalyclip: Object- agnostic prompt learning for zero-shot anomaly detection,","venue":null,"work_id":"93ac9fc5-c8c9-4ffe-821e-79ea49511e20","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":62,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.905039Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:108fc6266fdc5dd125d6f842b11b0f625b7b529654fa206ef82b7fccbf118507","observation_id":"1e0c80f1-bbe7-421e-8bdd-1804c835419b","resolution":{"observed_at":"2026-08-05T14:01:43.045923Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.837408Z","title":"Adapting visual-language models for generalizable anomaly detection in medical images,","venue":null,"work_id":"976e3c93-d84b-4f08-8caf-10ed1ad80f58","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":63,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:40.982042Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:3548981f5596b1b0ce47a8620bfa3567c41bb3ae6b1124bad368c4ae373e611c","observation_id":"a1375852-3841-417b-91b7-89e1fc25b1e8","resolution":{"observed_at":"2026-08-05T14:01:42.872108Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.710181Z","title":"Adaclip: Adapting clip with hybrid learnable prompts for zero-shot anomaly detection,","venue":null,"work_id":"15b72298-3cd7-4139-90f2-5a6105f776ee","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":64,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.070770Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:d447b449660bf3e2870f6161c6daec576abdb65152d293cfaef78101d54b0f6b","observation_id":"01ad627b-e531-4cb3-92b1-4b53fd5f7b17","resolution":{"observed_at":"2026-08-05T14:01:42.790919Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.606493Z","title":"Grounding dino: Marrying dino with grounded pre-training for open-set object detection,","venue":null,"work_id":"e8cc0112-b602-4b53-90a5-a93855d20a6c","year":2024},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":65,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.139609Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:d9df3e41f378af9dc90c84f83ab75197d8e7f4cdf60d01bab2043c21a0884449","observation_id":"681cdad2-9055-4bf3-829e-e3e865893f98","resolution":{"observed_at":"2026-08-05T14:01:42.646141Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.516981Z","title":"Simplenet: A simple network for image anomaly detection and localization,","venue":null,"work_id":"b86b17f4-ad0c-438c-bdee-9f750f6c8a9b","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":66,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.205554Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:dccc8d10f1ef4b01b50a540ffbb4f34ba383d9de493bc2acd99ea6c440628c8d","observation_id":"2f928faa-d09b-44c5-acf7-c3cc18b3c143","resolution":{"observed_at":"2026-08-05T14:01:42.564177Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.366653Z","title":"Exploring the limits of transfer learning with a unified text-to-text transformer,","venue":null,"work_id":"9ec36d8e-8d4a-40fe-a04f-c2c777882172","year":2020},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":67,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.253619Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:279a8b4fd23e87d5036e62f0d1623157e5d139f5442c052015bde077b7b6d7b8","observation_id":"43e60add-9ace-46cd-ae41-cf799628bacc","resolution":{"observed_at":"2026-08-05T14:01:42.459044Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.267201Z","title":"Detecting human- object contact in images,","venue":null,"work_id":"2f750499-907a-4505-be53-e4b21e6c4a79","year":2023},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":68,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.340093Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:de9866ef4f5b5fecc70217e85da412c3aa57d2e7b6b6f067e445e9605e8a343d","observation_id":"9a4d3ea3-5a4a-4701-a15f-da5edaf12f37","resolution":{"observed_at":"2026-08-05T14:01:42.310817Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.159119Z","title":"Interactvlm: 3d interaction reasoning from 2d foundational models,","venue":null,"work_id":"74d57cb4-5dc4-4f3d-9f93-7618e0122a2f","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":69,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.407294Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:8ca7b50e919ca5c3386ee86a649f7cd9b3c8f17da5fa55974a2dbc9ac06a5afa","observation_id":"ba0dcbe9-0a58-4ba7-93cf-0a0413a3ca34","resolution":{"observed_at":"2026-08-05T14:01:42.208577Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:42.037655Z","title":"Product quantization for nearest neighbor search,","venue":null,"work_id":"70520b9d-7eb7-4e6f-b4cd-0ddcec635039","year":2010},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":70,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.475995Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:e902584f9c76ef4e350618ce775aa8b76d0c7ac0dcbc109012ecccffdb089e99","observation_id":"d0ad3774-0c82-47d8-8e31-1c8964e9dbb3","resolution":{"observed_at":"2026-08-05T14:01:42.105552Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:41.756302Z","title":"Her research interests include deep learning on image processing and medical image processing and applications","venue":null,"work_id":"cdd25bcb-0d5e-408b-a29f-44df79921077","year":null},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":2014,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.595916Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:7e603da20f9913ab685b671b45f6b05c4c81d1eac1fca3aab3806e91ffa93793","observation_id":"90732f3f-36cd-4b95-a434-a695a5c2aae5","resolution":{"observed_at":"2026-08-05T14:01:41.779090Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}},{"citation":{"cited_paper":null,"cited_work":{"arxiv_id":null,"doi":null,"metadata_source":"raw_reference","pith_arxiv_id":null,"snapshot_observed_at":"2026-08-05T14:01:41.874445Z","title":"His research interests in- clude deep learning, illumination processing, image restoration, shadow removal, anomaly detection, and diffusion models","venue":null,"work_id":"e9b4ac81-0d78-4f7b-be9f-64fad308e284","year":2025},"citing_paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank","version":1},"reference_index":2024,"source":"pdf_text","source_observed_at":"2026-08-05T14:01:41.533436Z"},"links":{"citing_paper":"/paper/2508.21795"},"observation_digest":"sha256:1d10f7ba1d1e8acc8eb696599b6ce3e89d4f6d0270879d306623cece7687b36a","observation_id":"deaf45ed-0c2f-411e-ab6b-34943de3eb80","resolution":{"observed_at":"2026-08-05T14:01:41.969727Z","resolver_source":"raw_fallback","status":"verified_fuzzy"},"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"}}],"paper":{"arxiv_id":"2508.21795","last_updated":"2025-08-29T17:22:13Z","latest_version":1,"primary_category":"cs.CV","snapshot_observed_at":"2026-08-09T15:16:10.024553Z","submitted_at":"2025-08-29T17:22:13Z","title":"TMUAD: Enhancing Logical Capabilities in Unified Anomaly Detection Models with a Text Memory Bank"},"reference_resolution":{"displayed":72,"state_counts":{"malformed_identifier":0,"metadata_mismatch":0,"parse_uncertain":0,"unresolved":6,"verified_exact":0,"verified_fuzzy":66},"total_outbound_references":72},"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 72 of 72 outbound references and 0 inbound Pith citation observations for arXiv:2508.21795."}