{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ","short_pith_number":"pith:ZLLB2PNE","schema_version":"1.0","canonical_sha256":"cad61d3da4c85c4d653e6f9a7b37b546447d59e3a4b229d7346162af6de123d1","source":{"kind":"arxiv","id":"2511.07403","version":2},"attestation_state":"computed","paper":{"title":"SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Cihang Xie, Haoqin Tu, Hardy Chen, Hunar Batra, Ronald Clark, Yuanze Lin","submitted_at":"2025-11-10T18:52:47Z","abstract_excerpt":"Multimodal large language models (MLLMs) have achieved remarkable progress in vision-language tasks, but continue to struggle with spatial reasoning. Existing spatial MLLMs rely on large-scale datasets, explicit 3D inputs, architecture-specific modifications, or sparse Reinforcement Learning (RL) methods that provide insufficient guidance for spatially-grounded reasoning. We introduce SpatialThinker. To our knowledge, it is the first MLLM unifying Scene Graph Generation (SGG) and visual reasoning in a single pass via online RL. The model simulates human-like spatial perception by constructing "},"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":"2511.07403","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-11-10T18:52:47Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"c2b6382f6cad79de25fc995fe985e8bbf76aea05fea0e3ec68ccc6fc33d13c05","abstract_canon_sha256":"625c64106af67e10e66bc09d73f4dca33e8e4cf5b725d7d796ff17fde9902b4b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T00:15:50.676219Z","signature_b64":"uSLoxRREtEz4H24xhlOdJs+EUdfsTWb1L1OQg3KuMNTLzQv6qZPFOiZmK6qW9suY19VUi5h5yNDd/RDoSRNSAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cad61d3da4c85c4d653e6f9a7b37b546447d59e3a4b229d7346162af6de123d1","last_reissued_at":"2026-07-07T00:15:50.674997Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T00:15:50.674997Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SpatialThinker: Reinforcing Scene Graph-Grounded Spatial Reasoning via Dense Rewards","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Cihang Xie, Haoqin Tu, Hardy Chen, Hunar Batra, Ronald Clark, Yuanze Lin","submitted_at":"2025-11-10T18:52:47Z","abstract_excerpt":"Multimodal large language models (MLLMs) have achieved remarkable progress in vision-language tasks, but continue to struggle with spatial reasoning. Existing spatial MLLMs rely on large-scale datasets, explicit 3D inputs, architecture-specific modifications, or sparse Reinforcement Learning (RL) methods that provide insufficient guidance for spatially-grounded reasoning. We introduce SpatialThinker. To our knowledge, it is the first MLLM unifying Scene Graph Generation (SGG) and visual reasoning in a single pass via online RL. The model simulates human-like spatial perception by constructing "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.07403","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/2511.07403/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":"2511.07403","created_at":"2026-07-07T00:15:50.675156+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.07403v2","created_at":"2026-07-07T00:15:50.675156+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.07403","created_at":"2026-07-07T00:15:50.675156+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZLLB2PNEZBOE","created_at":"2026-07-07T00:15:50.675156+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZLLB2PNEZBOE2ZJ6","created_at":"2026-07-07T00:15:50.675156+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZLLB2PNE","created_at":"2026-07-07T00:15:50.675156+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":10,"sample":[{"citing_arxiv_id":"2604.21190","citing_title":"SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19253","citing_title":"OneCanvas: 3D Scene Understanding via Panoramic Reprojection","ref_index":2,"is_internal_anchor":true},{"citing_arxiv_id":"2607.00881","citing_title":"OmniView-Space: Reinforcing Spatial Reasoning via Multi-Perspective Spatial Mapping","ref_index":58,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23176","citing_title":"DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving","ref_index":95,"is_internal_anchor":true},{"citing_arxiv_id":"2605.25802","citing_title":"Rethinking VLM Representation for VLA Initialization","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.23176","citing_title":"DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving","ref_index":95,"is_internal_anchor":true},{"citing_arxiv_id":"2505.17012","citing_title":"SpatialScore: Towards Comprehensive Evaluation for Spatial Intelligence","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2605.10106","citing_title":"ViSRA: A Video-based Spatial Reasoning Agent for Multi-modal Large Language Models","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.07148","citing_title":"Uncovering and Shaping the Latent Representation of 3D Scene Topology in Vision-Language Models","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2604.21190","citing_title":"SpatiO: Adaptive Test-Time Orchestration of Vision-Language Agents for Spatial Reasoning","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ","json":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ.json","graph_json":"https://pith.science/api/pith-number/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/graph.json","events_json":"https://pith.science/api/pith-number/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/events.json","paper":"https://pith.science/paper/ZLLB2PNE"},"agent_actions":{"view_html":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ","download_json":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ.json","view_paper":"https://pith.science/paper/ZLLB2PNE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.07403&json=true","fetch_graph":"https://pith.science/api/pith-number/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/graph.json","fetch_events":"https://pith.science/api/pith-number/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/action/storage_attestation","attest_author":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/action/author_attestation","sign_citation":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/action/citation_signature","submit_replication":"https://pith.science/pith/ZLLB2PNEZBOE2ZJ6N6NHWN5VIZ/action/replication_record"}},"created_at":"2026-07-07T00:15:50.675156+00:00","updated_at":"2026-07-07T00:15:50.675156+00:00"}