{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:SYHSW3NXWPPSNFAOUVIQST2RBG","short_pith_number":"pith:SYHSW3NX","schema_version":"1.0","canonical_sha256":"960f2b6db7b3df26940ea551094f5109856b782cfd794712441f4603ce76403a","source":{"kind":"arxiv","id":"2412.17451","version":3},"attestation_state":"computed","paper":{"title":"Diving into Self-Evolving Training for Multimodal Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fan Zhou, Junlong Li, Junxian He, Wei Liu, Xiwen Zhang, Yu Cheng","submitted_at":"2024-12-23T10:18:41Z","abstract_excerpt":"Self-evolving trainin--where models iteratively learn from their own outputs--has emerged as a key approach for complex reasoning tasks, addressing the scarcity of high-quality chain-of-thought data. However, its effectiveness in multimodal reasoning, a domain more intricate than text-only reasoning, remains underexplored, and the understanding of critical factors in this training paradigm remains limited. Furthermore, a central challenge for this training method is performance saturation, which impedes further improvements and scalability. Inspired by reinforcement learning (RL), in this pape"},"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":"2412.17451","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-12-23T10:18:41Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG"],"title_canon_sha256":"a000f3ebdfbe1a449d02d16fb308d7d4c17fa284f882dee50852fe5a1b15dcce","abstract_canon_sha256":"3575fca06d059db11e4614086e1fa8ac935f26918b0207bbef27f0bb72d67117"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:16:53.583038Z","signature_b64":"BftDZ1krijopJSn20+R0RDxfOIVmXNiVwmqr4qjef3LeUxpgWjW7OPjCJPh66zpNiy/bBRxVSh/9e9umxoCwBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"960f2b6db7b3df26940ea551094f5109856b782cfd794712441f4603ce76403a","last_reissued_at":"2026-07-05T11:16:53.582545Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:16:53.582545Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Diving into Self-Evolving Training for Multimodal Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG"],"primary_cat":"cs.CL","authors_text":"Fan Zhou, Junlong Li, Junxian He, Wei Liu, Xiwen Zhang, Yu Cheng","submitted_at":"2024-12-23T10:18:41Z","abstract_excerpt":"Self-evolving trainin--where models iteratively learn from their own outputs--has emerged as a key approach for complex reasoning tasks, addressing the scarcity of high-quality chain-of-thought data. However, its effectiveness in multimodal reasoning, a domain more intricate than text-only reasoning, remains underexplored, and the understanding of critical factors in this training paradigm remains limited. Furthermore, a central challenge for this training method is performance saturation, which impedes further improvements and scalability. Inspired by reinforcement learning (RL), in this pape"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.17451","kind":"arxiv","version":3},"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/2412.17451/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":"2412.17451","created_at":"2026-07-05T11:16:53.582603+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.17451v3","created_at":"2026-07-05T11:16:53.582603+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.17451","created_at":"2026-07-05T11:16:53.582603+00:00"},{"alias_kind":"pith_short_12","alias_value":"SYHSW3NXWPPS","created_at":"2026-07-05T11:16:53.582603+00:00"},{"alias_kind":"pith_short_16","alias_value":"SYHSW3NXWPPSNFAO","created_at":"2026-07-05T11:16:53.582603+00:00"},{"alias_kind":"pith_short_8","alias_value":"SYHSW3NX","created_at":"2026-07-05T11:16:53.582603+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.27373","citing_title":"Paying More Attention to Visual Tokens in Self-Evolving Large Multimodal Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2606.11470","citing_title":"The Periodic Table of LLM Reasoning: A Structured Survey of Reasoning Paradigms, Methods, and Failure Modes","ref_index":156,"is_internal_anchor":false},{"citing_arxiv_id":"2503.12605","citing_title":"Multimodal Chain-of-Thought Reasoning: A Comprehensive Survey","ref_index":241,"is_internal_anchor":false},{"citing_arxiv_id":"2502.17419","citing_title":"From System 1 to System 2: A Survey of Reasoning Large Language Models","ref_index":176,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18320","citing_title":"EVE: Verifiable Self-Evolution of MLLMs via Executable Visual Transformations","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01048","citing_title":"Compared to What? Baselines and Metrics for Counterfactual Prompting","ref_index":71,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG","json":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG.json","graph_json":"https://pith.science/api/pith-number/SYHSW3NXWPPSNFAOUVIQST2RBG/graph.json","events_json":"https://pith.science/api/pith-number/SYHSW3NXWPPSNFAOUVIQST2RBG/events.json","paper":"https://pith.science/paper/SYHSW3NX"},"agent_actions":{"view_html":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG","download_json":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG.json","view_paper":"https://pith.science/paper/SYHSW3NX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.17451&json=true","fetch_graph":"https://pith.science/api/pith-number/SYHSW3NXWPPSNFAOUVIQST2RBG/graph.json","fetch_events":"https://pith.science/api/pith-number/SYHSW3NXWPPSNFAOUVIQST2RBG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG/action/storage_attestation","attest_author":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG/action/author_attestation","sign_citation":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG/action/citation_signature","submit_replication":"https://pith.science/pith/SYHSW3NXWPPSNFAOUVIQST2RBG/action/replication_record"}},"created_at":"2026-07-05T11:16:53.582603+00:00","updated_at":"2026-07-05T11:16:53.582603+00:00"}