{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CF5MIOSGXMF5BJXSP4RLLN7TE7","short_pith_number":"pith:CF5MIOSG","schema_version":"1.0","canonical_sha256":"117ac43a46bb0bd0a6f27f22b5b7f327dca2bb869387a7d0bb04787f2fc11614","source":{"kind":"arxiv","id":"2504.16072","version":1},"attestation_state":"computed","paper":{"title":"Describe Anything: Detailed Localized Image and Video Captioning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Adam Yala, Boyi Li, Hanzi Mao, Long Lian, Marco Pavone, Ming-Yu Liu, Sifei Liu, Trevor Darrell, Yifan Ding, Yin Cui, Yunhao Ge","submitted_at":"2025-04-22T17:51:41Z","abstract_excerpt":"Generating detailed and accurate descriptions for specific regions in images and videos remains a fundamental challenge for vision-language models. We introduce the Describe Anything Model (DAM), a model designed for detailed localized captioning (DLC). DAM preserves both local details and global context through two key innovations: a focal prompt, which ensures high-resolution encoding of targeted regions, and a localized vision backbone, which integrates precise localization with its broader context. To tackle the scarcity of high-quality DLC data, we propose a Semi-supervised learning (SSL)"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2504.16072","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-22T17:51:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"296424ccf4248476dc53dd220e8e22c5b445e497f764b52938460bb92b2cc12e","abstract_canon_sha256":"71a909d979c9815ca7b132cb907a312a8fd65d0d2b82e64f3affb060606df25e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:52:35.041244Z","signature_b64":"qSSK/W87QpSOvEhzC+KCoIlot3j54Swjaowp9GvjGHjj5+WMuOCw5Ka4FvoGjDMoxA55cW6hFb9HEEAu4xYLAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"117ac43a46bb0bd0a6f27f22b5b7f327dca2bb869387a7d0bb04787f2fc11614","last_reissued_at":"2026-07-05T10:52:35.040787Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:52:35.040787Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Describe Anything: Detailed Localized Image and Video Captioning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Adam Yala, Boyi Li, Hanzi Mao, Long Lian, Marco Pavone, Ming-Yu Liu, Sifei Liu, Trevor Darrell, Yifan Ding, Yin Cui, Yunhao Ge","submitted_at":"2025-04-22T17:51:41Z","abstract_excerpt":"Generating detailed and accurate descriptions for specific regions in images and videos remains a fundamental challenge for vision-language models. We introduce the Describe Anything Model (DAM), a model designed for detailed localized captioning (DLC). DAM preserves both local details and global context through two key innovations: a focal prompt, which ensures high-resolution encoding of targeted regions, and a localized vision backbone, which integrates precise localization with its broader context. To tackle the scarcity of high-quality DLC data, we propose a Semi-supervised learning (SSL)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.16072","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2504.16072/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2504.16072","created_at":"2026-07-05T10:52:35.040844+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.16072v1","created_at":"2026-07-05T10:52:35.040844+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.16072","created_at":"2026-07-05T10:52:35.040844+00:00"},{"alias_kind":"pith_short_12","alias_value":"CF5MIOSGXMF5","created_at":"2026-07-05T10:52:35.040844+00:00"},{"alias_kind":"pith_short_16","alias_value":"CF5MIOSGXMF5BJXS","created_at":"2026-07-05T10:52:35.040844+00:00"},{"alias_kind":"pith_short_8","alias_value":"CF5MIOSG","created_at":"2026-07-05T10:52:35.040844+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":12,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05348","citing_title":"Beyond Isolated Objects: Relationship-aware Open Vocabulary Scene Understanding via 3D Scene Graph Analysis","ref_index":30,"is_internal_anchor":true},{"citing_arxiv_id":"2606.15476","citing_title":"FARM: Find Anything using Relational Spatial Memory","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2606.05031","citing_title":"MetaPoint: Unlocking Precise Spatial Control in Agentic Visual Generation","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20278","citing_title":"ClaimDiff-RL: Fine-Grained Caption Reinforcement Learning through Visual Claim Comparison","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20278","citing_title":"ClaimDiff-RL: Fine-Grained Caption Reinforcement Learning through Visual Claim Comparison","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.18018","citing_title":"See What I Mean: Aligning Vision and Language Representations for Video Fine-grained Object Understanding","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20165","citing_title":"CaMo: Camera Motion Grounded Evaluation and Training for Vision-Language Models","ref_index":77,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16903","citing_title":"WOW-Seg: A Word-free Open World Segmentation Model","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2511.21025","citing_title":"CaptionQA: Is Your Caption as Useful as the Image Itself?","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.01907","citing_title":"Lifting Unlabeled Internet-level Data for 3D Scene Understanding","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"2605.12305","citing_title":"Images in Sentences: Scaling Interleaved Instructions for Unified Visual Generation","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21718","citing_title":"Building a Precise Video Language with Human-AI Oversight","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7","json":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7.json","graph_json":"https://pith.science/api/pith-number/CF5MIOSGXMF5BJXSP4RLLN7TE7/graph.json","events_json":"https://pith.science/api/pith-number/CF5MIOSGXMF5BJXSP4RLLN7TE7/events.json","paper":"https://pith.science/paper/CF5MIOSG"},"agent_actions":{"view_html":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7","download_json":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7.json","view_paper":"https://pith.science/paper/CF5MIOSG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.16072&json=true","fetch_graph":"https://pith.science/api/pith-number/CF5MIOSGXMF5BJXSP4RLLN7TE7/graph.json","fetch_events":"https://pith.science/api/pith-number/CF5MIOSGXMF5BJXSP4RLLN7TE7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7/action/storage_attestation","attest_author":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7/action/author_attestation","sign_citation":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7/action/citation_signature","submit_replication":"https://pith.science/pith/CF5MIOSGXMF5BJXSP4RLLN7TE7/action/replication_record"}},"created_at":"2026-07-05T10:52:35.040844+00:00","updated_at":"2026-07-05T10:52:35.040844+00:00"}