{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:B4RITHJEMG7NP2FMXKNG3SOIC4","short_pith_number":"pith:B4RITHJE","schema_version":"1.0","canonical_sha256":"0f22899d2461bed7e8acba9a6dc9c8170d3ac6ba0075f377ae798837f495e5a3","source":{"kind":"arxiv","id":"2504.09641","version":1},"attestation_state":"computed","paper":{"title":"TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lei Huang, Siwei Wen, Wenjun Wu, Xingjian Zhang","submitted_at":"2025-04-13T16:32:49Z","abstract_excerpt":"Recently, improving the reasoning ability of large multimodal models (LMMs) through reinforcement learning has made great progress. However, most existing works are based on highly reasoning-intensive datasets such as mathematics and code, and researchers generally choose large-scale models as the foundation. We argue that exploring small-scale models' reasoning capabilities remains valuable for researchers with limited computational resources. Moreover, enabling models to explain their reasoning processes on general question-answering datasets is equally meaningful. Therefore, we present the "},"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":"2504.09641","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-04-13T16:32:49Z","cross_cats_sorted":[],"title_canon_sha256":"60f8ec5c17587e2cfa3d7929e2c3a4c29b42666da56dd3421d11848afe2eba2a","abstract_canon_sha256":"ca51ddaf0a5eebb5c76fbdacee9dbc50ab36c75c644bf7516979739fb42d56a2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:48:40.566996Z","signature_b64":"kbZ/CVl4kA3b0RfDYBxzDRLb4Tc2HcaXmdDnlWWQvKd11M2WAG5Cj9X8PsL49i3WMzu/LB7ow1J2K2vDu0qVCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f22899d2461bed7e8acba9a6dc9c8170d3ac6ba0075f377ae798837f495e5a3","last_reissued_at":"2026-07-05T10:48:40.566493Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:48:40.566493Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TinyLLaVA-Video-R1: Towards Smaller LMMs for Video Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lei Huang, Siwei Wen, Wenjun Wu, Xingjian Zhang","submitted_at":"2025-04-13T16:32:49Z","abstract_excerpt":"Recently, improving the reasoning ability of large multimodal models (LMMs) through reinforcement learning has made great progress. However, most existing works are based on highly reasoning-intensive datasets such as mathematics and code, and researchers generally choose large-scale models as the foundation. We argue that exploring small-scale models' reasoning capabilities remains valuable for researchers with limited computational resources. Moreover, enabling models to explain their reasoning processes on general question-answering datasets is equally meaningful. Therefore, we present the "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.09641","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/2504.09641/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":"2504.09641","created_at":"2026-07-05T10:48:40.566556+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.09641v1","created_at":"2026-07-05T10:48:40.566556+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.09641","created_at":"2026-07-05T10:48:40.566556+00:00"},{"alias_kind":"pith_short_12","alias_value":"B4RITHJEMG7N","created_at":"2026-07-05T10:48:40.566556+00:00"},{"alias_kind":"pith_short_16","alias_value":"B4RITHJEMG7NP2FM","created_at":"2026-07-05T10:48:40.566556+00:00"},{"alias_kind":"pith_short_8","alias_value":"B4RITHJE","created_at":"2026-07-05T10:48:40.566556+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19927","citing_title":"CARE: Competence-Aware Reward Shaping for Adaptive Reasoning Length in Video-MLLMs","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23216","citing_title":"CaST-Bench: Benchmarking Causal Chain-Grounded Spatio-Temporal Reasoning for Video Question Answering","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2606.18441","citing_title":"Reasoning as Intersection: Consensus-Frame Alignment for Visual Focus in Video-MLLMs","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2605.23216","citing_title":"CaST-Bench: Benchmarking Causal Chain-Grounded Spatio-Temporal Reasoning for Video Question Answering","ref_index":47,"is_internal_anchor":false},{"citing_arxiv_id":"2505.20715","citing_title":"MUSEG: Reinforcing Video Temporal Understanding via Timestamp-Aware Multi-Segment Grounding","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2507.00748","citing_title":"Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning","ref_index":52,"is_internal_anchor":false},{"citing_arxiv_id":"2511.11113","citing_title":"VIDEOP2R: Video Understanding from Perception to Reasoning","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2505.21374","citing_title":"Video-Holmes: Can MLLM Think Like Holmes for Complex Video Reasoning?","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2503.13377","citing_title":"Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2512.03963","citing_title":"TempR1: Improving Temporal Understanding of MLLMs via Temporal-Aware Multi-Task Reinforcement Learning","ref_index":70,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01324","citing_title":"Beyond Perceptual Shortcuts: Causal-Inspired Debiasing Optimization for Generalizable Video Reasoning in Lightweight MLLMs","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2604.20473","citing_title":"Video-ToC: Video Tree-of-Cue Reasoning","ref_index":42,"is_internal_anchor":false},{"citing_arxiv_id":"2604.04379","citing_title":"Reinforce to Learn, Elect to Reason: A Dual Paradigm for Video Reasoning","ref_index":55,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4","json":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4.json","graph_json":"https://pith.science/api/pith-number/B4RITHJEMG7NP2FMXKNG3SOIC4/graph.json","events_json":"https://pith.science/api/pith-number/B4RITHJEMG7NP2FMXKNG3SOIC4/events.json","paper":"https://pith.science/paper/B4RITHJE"},"agent_actions":{"view_html":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4","download_json":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4.json","view_paper":"https://pith.science/paper/B4RITHJE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.09641&json=true","fetch_graph":"https://pith.science/api/pith-number/B4RITHJEMG7NP2FMXKNG3SOIC4/graph.json","fetch_events":"https://pith.science/api/pith-number/B4RITHJEMG7NP2FMXKNG3SOIC4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4/action/storage_attestation","attest_author":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4/action/author_attestation","sign_citation":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4/action/citation_signature","submit_replication":"https://pith.science/pith/B4RITHJEMG7NP2FMXKNG3SOIC4/action/replication_record"}},"created_at":"2026-07-05T10:48:40.566556+00:00","updated_at":"2026-07-05T10:48:40.566556+00:00"}