{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OJAM2Q6JD2YRHFC2VZNNYKV6BO","short_pith_number":"pith:OJAM2Q6J","schema_version":"1.0","canonical_sha256":"7240cd43c91eb113945aae5adc2abe0bbe2e34db9c4c128bd6c9d94066e784ca","source":{"kind":"arxiv","id":"2502.14044","version":2},"attestation_state":"computed","paper":{"title":"Enhancing Cognition and Explainability of Multimodal Foundation Models with Self-Synthesized Data","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jin Sun, Ninghao Liu, Quanzheng Li, Xiang Li, Yucheng Shi","submitted_at":"2025-02-19T19:05:45Z","abstract_excerpt":"Large Multimodal Models (LMMs), or Vision-Language Models (VLMs), have shown impressive capabilities in a wide range of visual tasks. However, they often struggle with fine-grained visual reasoning, failing to identify domain-specific objectives and provide justifiable explanations for their predictions. To address the above challenge, we propose a novel visual rejection sampling framework to improve the cognition and explainability of LMMs using self-synthesized data. Specifically, visual fine-tuning requires images, queries, and target answers. Our approach begins by synthesizing interpretab"},"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":"2502.14044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-02-19T19:05:45Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"2926e39d81b928e0be047e2c71e81ba53af7d47595da3f02e37a85c6b19cf6cc","abstract_canon_sha256":"0d3af62990cb33ea3d88104542d002a7a797a23281df2d19d2a3dbd23779890a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:19:39.643601Z","signature_b64":"3WU2vyZ4LT8hjhFWaUBfZcnsLrrk3ioWO7wEc0kcmdl7/ZHoZ9Mxc+4J04VpCyTJjVw0CGTQlVV3AvQuaf+ZAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7240cd43c91eb113945aae5adc2abe0bbe2e34db9c4c128bd6c9d94066e784ca","last_reissued_at":"2026-07-05T10:19:39.643102Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:19:39.643102Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Enhancing Cognition and Explainability of Multimodal Foundation Models with Self-Synthesized Data","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Jin Sun, Ninghao Liu, Quanzheng Li, Xiang Li, Yucheng Shi","submitted_at":"2025-02-19T19:05:45Z","abstract_excerpt":"Large Multimodal Models (LMMs), or Vision-Language Models (VLMs), have shown impressive capabilities in a wide range of visual tasks. However, they often struggle with fine-grained visual reasoning, failing to identify domain-specific objectives and provide justifiable explanations for their predictions. To address the above challenge, we propose a novel visual rejection sampling framework to improve the cognition and explainability of LMMs using self-synthesized data. Specifically, visual fine-tuning requires images, queries, and target answers. Our approach begins by synthesizing interpretab"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.14044","kind":"arxiv","version":2},"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/2502.14044/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":"2502.14044","created_at":"2026-07-05T10:19:39.643162+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.14044v2","created_at":"2026-07-05T10:19:39.643162+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.14044","created_at":"2026-07-05T10:19:39.643162+00:00"},{"alias_kind":"pith_short_12","alias_value":"OJAM2Q6JD2YR","created_at":"2026-07-05T10:19:39.643162+00:00"},{"alias_kind":"pith_short_16","alias_value":"OJAM2Q6JD2YRHFC2","created_at":"2026-07-05T10:19:39.643162+00:00"},{"alias_kind":"pith_short_8","alias_value":"OJAM2Q6J","created_at":"2026-07-05T10:19:39.643162+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01612","citing_title":"Self-Improving Small Object Grounding in LVLMs","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2602.07605","citing_title":"Fine-R1: Make Multi-modal LLMs Excel in Fine-Grained Visual Recognition by Chain-of-Thought Reasoning","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO","json":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO.json","graph_json":"https://pith.science/api/pith-number/OJAM2Q6JD2YRHFC2VZNNYKV6BO/graph.json","events_json":"https://pith.science/api/pith-number/OJAM2Q6JD2YRHFC2VZNNYKV6BO/events.json","paper":"https://pith.science/paper/OJAM2Q6J"},"agent_actions":{"view_html":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO","download_json":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO.json","view_paper":"https://pith.science/paper/OJAM2Q6J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.14044&json=true","fetch_graph":"https://pith.science/api/pith-number/OJAM2Q6JD2YRHFC2VZNNYKV6BO/graph.json","fetch_events":"https://pith.science/api/pith-number/OJAM2Q6JD2YRHFC2VZNNYKV6BO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO/action/storage_attestation","attest_author":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO/action/author_attestation","sign_citation":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO/action/citation_signature","submit_replication":"https://pith.science/pith/OJAM2Q6JD2YRHFC2VZNNYKV6BO/action/replication_record"}},"created_at":"2026-07-05T10:19:39.643162+00:00","updated_at":"2026-07-05T10:19:39.643162+00:00"}