{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:WIFEHNJ2RCB34D3DQUZSPVBU2J","short_pith_number":"pith:WIFEHNJ2","schema_version":"1.0","canonical_sha256":"b20a43b53a8883be0f63853327d434d277e1175e5acc1f177b455a72c54bebb7","source":{"kind":"arxiv","id":"2506.05384","version":2},"attestation_state":"computed","paper":{"title":"Q-Ponder: A Unified Training Pipeline for Reasoning-based Visual Quality Assessment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Bo Li, Bowen Tang, Jian Zhang, Jinwen Chen, Lujian Yao, Peng-Tao Jiang, Qiyuan Wang, Wenxiang Chen, Xinbin Yuan, Zhuoxuan Cai","submitted_at":"2025-06-03T10:11:51Z","abstract_excerpt":"Recent studies demonstrate that multimodal large language models (MLLMs) can proficiently evaluate visual quality through interpretable assessments. However, existing approaches typically treat quality scoring and reasoning descriptions as separate tasks with disjoint optimization objectives, leading to a trade-off: models adept at quality reasoning descriptions struggle with precise score regression, while score-focused models lack interpretability. This limitation hinders the full potential of MLLMs in visual quality assessment, where accuracy and interpretability should be mutually reinforc"},"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":"2506.05384","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-03T10:11:51Z","cross_cats_sorted":["cs.AI","cs.MM"],"title_canon_sha256":"5c3e45df5317f0ed1b3fc8be695fb1270b60093b8fa3661ed0b699018ae97073","abstract_canon_sha256":"6d22433269997e7625560d54cedda729c227ef0652f86d8e331607ee1a3fadd2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:38.441852Z","signature_b64":"MmpnwPczTKZw4YXgubbG5Bjx70jS9kEqr0LTEoMsoPothl5SnPutVJA6BQmHhlbBkI64OQUR/fxzmf9me9TsCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b20a43b53a8883be0f63853327d434d277e1175e5acc1f177b455a72c54bebb7","last_reissued_at":"2026-07-05T11:20:38.441386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:38.441386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Q-Ponder: A Unified Training Pipeline for Reasoning-based Visual Quality Assessment","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.MM"],"primary_cat":"cs.CV","authors_text":"Bo Li, Bowen Tang, Jian Zhang, Jinwen Chen, Lujian Yao, Peng-Tao Jiang, Qiyuan Wang, Wenxiang Chen, Xinbin Yuan, Zhuoxuan Cai","submitted_at":"2025-06-03T10:11:51Z","abstract_excerpt":"Recent studies demonstrate that multimodal large language models (MLLMs) can proficiently evaluate visual quality through interpretable assessments. However, existing approaches typically treat quality scoring and reasoning descriptions as separate tasks with disjoint optimization objectives, leading to a trade-off: models adept at quality reasoning descriptions struggle with precise score regression, while score-focused models lack interpretability. This limitation hinders the full potential of MLLMs in visual quality assessment, where accuracy and interpretability should be mutually reinforc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.05384","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/2506.05384/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":"2506.05384","created_at":"2026-07-05T11:20:38.441445+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.05384v2","created_at":"2026-07-05T11:20:38.441445+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.05384","created_at":"2026-07-05T11:20:38.441445+00:00"},{"alias_kind":"pith_short_12","alias_value":"WIFEHNJ2RCB3","created_at":"2026-07-05T11:20:38.441445+00:00"},{"alias_kind":"pith_short_16","alias_value":"WIFEHNJ2RCB34D3D","created_at":"2026-07-05T11:20:38.441445+00:00"},{"alias_kind":"pith_short_8","alias_value":"WIFEHNJ2","created_at":"2026-07-05T11:20:38.441445+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.29760","citing_title":"MR-IQA: A Unified Margin View of Regression and Ranking for Blind Image Quality Assessment","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2606.29760","citing_title":"MR-IQA: A Unified Margin View of Regression and Ranking for Blind Image Quality Assessment","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2601.15356","citing_title":"Q-Probe: Scaling Image Quality Assessment to High Resolution via Context-Aware Agentic Probing","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19587","citing_title":"SmartPhotoCrafter: Unified Reasoning, Generation and Optimization for Automatic Photographic Image Editing","ref_index":5,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J","json":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J.json","graph_json":"https://pith.science/api/pith-number/WIFEHNJ2RCB34D3DQUZSPVBU2J/graph.json","events_json":"https://pith.science/api/pith-number/WIFEHNJ2RCB34D3DQUZSPVBU2J/events.json","paper":"https://pith.science/paper/WIFEHNJ2"},"agent_actions":{"view_html":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J","download_json":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J.json","view_paper":"https://pith.science/paper/WIFEHNJ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.05384&json=true","fetch_graph":"https://pith.science/api/pith-number/WIFEHNJ2RCB34D3DQUZSPVBU2J/graph.json","fetch_events":"https://pith.science/api/pith-number/WIFEHNJ2RCB34D3DQUZSPVBU2J/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J/action/storage_attestation","attest_author":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J/action/author_attestation","sign_citation":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J/action/citation_signature","submit_replication":"https://pith.science/pith/WIFEHNJ2RCB34D3DQUZSPVBU2J/action/replication_record"}},"created_at":"2026-07-05T11:20:38.441445+00:00","updated_at":"2026-07-05T11:20:38.441445+00:00"}