{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:3BE4J2GE7ZIYPD3LZ7IQABGZUO","short_pith_number":"pith:3BE4J2GE","schema_version":"1.0","canonical_sha256":"d849c4e8c4fe51878f6bcfd10004d9a3a8bcc2f150eb2ff1ea04d6f270c4f3f1","source":{"kind":"arxiv","id":"2506.17545","version":1},"attestation_state":"computed","paper":{"title":"Scene-R1: Video-Grounded Large Language Models for 3D Scene Reasoning without 3D Annotations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chun-Mei Feng, Na Zhao, Shuguang Cui, Shuyi Jiang, Yaolun Zhang, Zhen Li, Zhihao Yuan","submitted_at":"2025-06-21T02:29:10Z","abstract_excerpt":"Currently, utilizing large language models to understand the 3D world is becoming popular. Yet existing 3D-aware LLMs act as black boxes: they output bounding boxes or textual answers without revealing how those decisions are made, and they still rely on pre-trained 3D detectors to supply object proposals. We introduce Scene-R1, a video-grounded framework that learns to reason about 3D scenes without any point-wise 3D instance supervision by pairing reinforcement-learning-driven reasoning with a two-stage grounding pipeline. In the temporal grounding stage, we explicitly reason about the video"},"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":"2506.17545","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-06-21T02:29:10Z","cross_cats_sorted":[],"title_canon_sha256":"a26a558c0eb642acf826d0ec355b2ac5712fb1499c2f5da209d89c68d033c848","abstract_canon_sha256":"44fcb6b85e2d64030ff48464014838a5905795367b1da4d158894e82163591fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:25:19.326091Z","signature_b64":"I2oUi1vVw9Qrr5SQnahes2Qxjprd6Z/FWq//BYeNLIVjcLM2WJOnS3rpuPnrzWqgK3Z/laxGkS4lfg43YXvKDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d849c4e8c4fe51878f6bcfd10004d9a3a8bcc2f150eb2ff1ea04d6f270c4f3f1","last_reissued_at":"2026-07-05T11:25:19.325609Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:25:19.325609Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scene-R1: Video-Grounded Large Language Models for 3D Scene Reasoning without 3D Annotations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chun-Mei Feng, Na Zhao, Shuguang Cui, Shuyi Jiang, Yaolun Zhang, Zhen Li, Zhihao Yuan","submitted_at":"2025-06-21T02:29:10Z","abstract_excerpt":"Currently, utilizing large language models to understand the 3D world is becoming popular. Yet existing 3D-aware LLMs act as black boxes: they output bounding boxes or textual answers without revealing how those decisions are made, and they still rely on pre-trained 3D detectors to supply object proposals. We introduce Scene-R1, a video-grounded framework that learns to reason about 3D scenes without any point-wise 3D instance supervision by pairing reinforcement-learning-driven reasoning with a two-stage grounding pipeline. In the temporal grounding stage, we explicitly reason about the video"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.17545","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/2506.17545/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":"2506.17545","created_at":"2026-07-05T11:25:19.325666+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.17545v1","created_at":"2026-07-05T11:25:19.325666+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.17545","created_at":"2026-07-05T11:25:19.325666+00:00"},{"alias_kind":"pith_short_12","alias_value":"3BE4J2GE7ZIY","created_at":"2026-07-05T11:25:19.325666+00:00"},{"alias_kind":"pith_short_16","alias_value":"3BE4J2GE7ZIYPD3L","created_at":"2026-07-05T11:25:19.325666+00:00"},{"alias_kind":"pith_short_8","alias_value":"3BE4J2GE","created_at":"2026-07-05T11:25:19.325666+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.01215","citing_title":"Distilling Neuro-Symbolic Programs into 3D Multi-modal LLMs","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01215","citing_title":"Distilling Neuro-Symbolic Programs into 3D Multi-modal LLMs","ref_index":11,"is_internal_anchor":false},{"citing_arxiv_id":"2605.20165","citing_title":"CaMo: Camera Motion Grounded Evaluation and Training for Vision-Language Models","ref_index":71,"is_internal_anchor":false},{"citing_arxiv_id":"2604.02870","citing_title":"Token Warping Helps MLLMs Look from Nearby Viewpoints","ref_index":117,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO","json":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO.json","graph_json":"https://pith.science/api/pith-number/3BE4J2GE7ZIYPD3LZ7IQABGZUO/graph.json","events_json":"https://pith.science/api/pith-number/3BE4J2GE7ZIYPD3LZ7IQABGZUO/events.json","paper":"https://pith.science/paper/3BE4J2GE"},"agent_actions":{"view_html":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO","download_json":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO.json","view_paper":"https://pith.science/paper/3BE4J2GE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.17545&json=true","fetch_graph":"https://pith.science/api/pith-number/3BE4J2GE7ZIYPD3LZ7IQABGZUO/graph.json","fetch_events":"https://pith.science/api/pith-number/3BE4J2GE7ZIYPD3LZ7IQABGZUO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO/action/storage_attestation","attest_author":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO/action/author_attestation","sign_citation":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO/action/citation_signature","submit_replication":"https://pith.science/pith/3BE4J2GE7ZIYPD3LZ7IQABGZUO/action/replication_record"}},"created_at":"2026-07-05T11:25:19.325666+00:00","updated_at":"2026-07-05T11:25:19.325666+00:00"}