{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:UJ3YAYQKBNXRZWRY6RCD2XXU5R","short_pith_number":"pith:UJ3YAYQK","schema_version":"1.0","canonical_sha256":"a27780620a0b6f1cda38f4443d5ef4ec623b5baee62408e10a3f91198cdd3c1e","source":{"kind":"arxiv","id":"2501.12368","version":2},"attestation_state":"computed","paper":{"title":"InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Haodong Duan, Jiaqi Wang, Kai Chen, Pan Zhang, Shengyuan Ding, Shenxi Wu, Wenwei Zhang, Xiaoyi Dong, Yubo Ma, Yuhang Cao, Yuhang Zang, Ziyu Liu","submitted_at":"2025-01-21T18:47:32Z","abstract_excerpt":"Despite the promising performance of Large Vision Language Models (LVLMs) in visual understanding, they occasionally generate incorrect outputs. While reward models (RMs) with reinforcement learning or test-time scaling offer the potential for improving generation quality, a critical gap remains: publicly available multi-modal RMs for LVLMs are scarce, and the implementation details of proprietary models are often unclear. We bridge this gap with InternLM-XComposer2.5-Reward (IXC-2.5-Reward), a simple yet effective multi-modal reward model that aligns LVLMs with human preferences. To ensure th"},"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":"2501.12368","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-21T18:47:32Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"a80644a0b74fc5977c9af4d2f68d9e55fe983a7a97ba3acc255c40175e513eaa","abstract_canon_sha256":"84f4daad95c072fba2b01ae4cff9796414a388894cabae6acbdd7e8e66245112"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:05:38.675261Z","signature_b64":"aqoHySpMoJkoApputPomVxM2wMVGEYQ1wFQZ/OGuZCo1vTIZARzkMxsBVDX2QAdp/r6PG7hyD9FTiQewQkYVDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"a27780620a0b6f1cda38f4443d5ef4ec623b5baee62408e10a3f91198cdd3c1e","last_reissued_at":"2026-07-05T11:05:38.674729Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:05:38.674729Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"InternLM-XComposer2.5-Reward: A Simple Yet Effective Multi-Modal Reward Model","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Dahua Lin, Haodong Duan, Jiaqi Wang, Kai Chen, Pan Zhang, Shengyuan Ding, Shenxi Wu, Wenwei Zhang, Xiaoyi Dong, Yubo Ma, Yuhang Cao, Yuhang Zang, Ziyu Liu","submitted_at":"2025-01-21T18:47:32Z","abstract_excerpt":"Despite the promising performance of Large Vision Language Models (LVLMs) in visual understanding, they occasionally generate incorrect outputs. While reward models (RMs) with reinforcement learning or test-time scaling offer the potential for improving generation quality, a critical gap remains: publicly available multi-modal RMs for LVLMs are scarce, and the implementation details of proprietary models are often unclear. We bridge this gap with InternLM-XComposer2.5-Reward (IXC-2.5-Reward), a simple yet effective multi-modal reward model that aligns LVLMs with human preferences. To ensure th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.12368","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/2501.12368/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":"2501.12368","created_at":"2026-07-05T11:05:38.674791+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.12368v2","created_at":"2026-07-05T11:05:38.674791+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.12368","created_at":"2026-07-05T11:05:38.674791+00:00"},{"alias_kind":"pith_short_12","alias_value":"UJ3YAYQKBNXR","created_at":"2026-07-05T11:05:38.674791+00:00"},{"alias_kind":"pith_short_16","alias_value":"UJ3YAYQKBNXRZWRY","created_at":"2026-07-05T11:05:38.674791+00:00"},{"alias_kind":"pith_short_8","alias_value":"UJ3YAYQK","created_at":"2026-07-05T11:05:38.674791+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":7,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.08525","citing_title":"DriveReward: A Comprehensive Dataset and Generative Vision-Language Reward Model for Autonomous Driving","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2605.25661","citing_title":"DRM: Diffusion-based Reward Model With Step-wise Guidance","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2411.10440","citing_title":"LLaVA-CoT: Let Vision Language Models Reason Step-by-Step","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2603.13224","citing_title":"Visual-ERM: Reward Modeling for Visual Equivalence","ref_index":38,"is_internal_anchor":false},{"citing_arxiv_id":"2503.05236","citing_title":"Unified Reward Model for Multimodal Understanding and Generation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2503.01785","citing_title":"Visual-RFT: Visual Reinforcement Fine-Tuning","ref_index":45,"is_internal_anchor":false},{"citing_arxiv_id":"2504.07615","citing_title":"VLM-R1: A Stable and Generalizable R1-style Large Vision-Language Model","ref_index":56,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R","json":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R.json","graph_json":"https://pith.science/api/pith-number/UJ3YAYQKBNXRZWRY6RCD2XXU5R/graph.json","events_json":"https://pith.science/api/pith-number/UJ3YAYQKBNXRZWRY6RCD2XXU5R/events.json","paper":"https://pith.science/paper/UJ3YAYQK"},"agent_actions":{"view_html":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R","download_json":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R.json","view_paper":"https://pith.science/paper/UJ3YAYQK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.12368&json=true","fetch_graph":"https://pith.science/api/pith-number/UJ3YAYQKBNXRZWRY6RCD2XXU5R/graph.json","fetch_events":"https://pith.science/api/pith-number/UJ3YAYQKBNXRZWRY6RCD2XXU5R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R/action/storage_attestation","attest_author":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R/action/author_attestation","sign_citation":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R/action/citation_signature","submit_replication":"https://pith.science/pith/UJ3YAYQKBNXRZWRY6RCD2XXU5R/action/replication_record"}},"created_at":"2026-07-05T11:05:38.674791+00:00","updated_at":"2026-07-05T11:05:38.674791+00:00"}