{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:JRN2MVLBNLQNVM4ZOY4BVW4HFT","short_pith_number":"pith:JRN2MVLB","schema_version":"1.0","canonical_sha256":"4c5ba655616ae0dab39976381adb872cdb847d9c09f215873e83e83193590efe","source":{"kind":"arxiv","id":"2505.23764","version":3},"attestation_state":"computed","paper":{"title":"MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chenming Zhu, Dahua Lin, Haodong Duan, Jiangmiao Pang, Jingli Lin, Mo Li, Runsen Xu, Sihan Yang, Sizhe Yang, Tai Wang, Xiangyu Yue, Xiaochen Chen, Yiman Xie","submitted_at":"2025-05-29T17:59:52Z","abstract_excerpt":"Spatial intelligence is essential for multimodal large language models (MLLMs) operating in the complex physical world. Existing benchmarks, however, probe only single-image relations and thus fail to assess the multi-image spatial reasoning that real-world deployments demand. We introduce MMSI-Bench, a VQA benchmark dedicated to multi-image spatial intelligence. Six 3D-vision researchers spent more than 300 hours meticulously crafting 1,000 challenging, unambiguous multiple-choice questions from over 120,000 images, each paired with carefully designed distractors and a stepwise reasoning proc"},"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":"2505.23764","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-29T17:59:52Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"94988001d7336d0113534cc9ac6926c328da0f560ddc68a6cbf73338a08b4d84","abstract_canon_sha256":"59fe32fca2730a537c3fcad1e4a77145d5739fa7efca652252b4a685bdf57d0a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:02:26.712093Z","signature_b64":"VMOQlWAOaWALyemU9cDC+HwSOjpVeK6O7EF1mkTUao1eSwAYanYURjcnc8Sayyxpnn106MB+W3vV3S+irT/lDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4c5ba655616ae0dab39976381adb872cdb847d9c09f215873e83e83193590efe","last_reissued_at":"2026-05-26T01:02:26.711194Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:02:26.711194Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MMSI-Bench: A Benchmark for Multi-Image Spatial Intelligence","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Chenming Zhu, Dahua Lin, Haodong Duan, Jiangmiao Pang, Jingli Lin, Mo Li, Runsen Xu, Sihan Yang, Sizhe Yang, Tai Wang, Xiangyu Yue, Xiaochen Chen, Yiman Xie","submitted_at":"2025-05-29T17:59:52Z","abstract_excerpt":"Spatial intelligence is essential for multimodal large language models (MLLMs) operating in the complex physical world. Existing benchmarks, however, probe only single-image relations and thus fail to assess the multi-image spatial reasoning that real-world deployments demand. We introduce MMSI-Bench, a VQA benchmark dedicated to multi-image spatial intelligence. Six 3D-vision researchers spent more than 300 hours meticulously crafting 1,000 challenging, unambiguous multiple-choice questions from over 120,000 images, each paired with carefully designed distractors and a stepwise reasoning proc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23764","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/2505.23764/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":"2505.23764","created_at":"2026-05-26T01:02:26.711305+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.23764v3","created_at":"2026-05-26T01:02:26.711305+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23764","created_at":"2026-05-26T01:02:26.711305+00:00"},{"alias_kind":"pith_short_12","alias_value":"JRN2MVLBNLQN","created_at":"2026-05-26T01:02:26.711305+00:00"},{"alias_kind":"pith_short_16","alias_value":"JRN2MVLBNLQNVM4Z","created_at":"2026-05-26T01:02:26.711305+00:00"},{"alias_kind":"pith_short_8","alias_value":"JRN2MVLB","created_at":"2026-05-26T01:02:26.711305+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":19,"internal_anchor_count":19,"sample":[{"citing_arxiv_id":"2605.23176","citing_title":"DRIVESPATIAL: A Benchmark for Spatiotemporal Intelligence in VLMs for Autonomous Driving","ref_index":110,"is_internal_anchor":true},{"citing_arxiv_id":"2605.21642","citing_title":"Ablate-to-Validate: Are Vision-Language Models Really Using Continuous Thought Tokens?","ref_index":31,"is_internal_anchor":true},{"citing_arxiv_id":"2605.22819","citing_title":"Cambrian-P: Pose-Grounded Video Understanding","ref_index":112,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04128","citing_title":"JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation","ref_index":96,"is_internal_anchor":true},{"citing_arxiv_id":"2605.20306","citing_title":"WildRoadBench: A Wild Aerial Road-Damage Grounding Benchmark for Vision-Language Models and Autonomous Agents","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2510.04978","citing_title":"Aligning Perception, Reasoning, Modeling and Interaction: A Survey on Physical AI","ref_index":86,"is_internal_anchor":true},{"citing_arxiv_id":"2512.17012","citing_title":"4D-RGPT: Toward Region-level 4D Understanding via Perceptual Distillation","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2602.11635","citing_title":"Do MLLMs Really Understand Space? A Mathematical Reasoning Evaluation","ref_index":39,"is_internal_anchor":true},{"citing_arxiv_id":"2603.03944","citing_title":"SCP: Spatial Causal Prediction in Video","ref_index":56,"is_internal_anchor":true},{"citing_arxiv_id":"2603.04676","citing_title":"Decoding the Pulse of Reasoning VLMs in Multi-Image Understanding Tasks","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2603.20633","citing_title":"Seed1.8 Model Card: Towards Generalized Real-World Agency","ref_index":83,"is_internal_anchor":true},{"citing_arxiv_id":"2604.26934","citing_title":"World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoning","ref_index":57,"is_internal_anchor":true},{"citing_arxiv_id":"2604.27393","citing_title":"MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2604.10506","citing_title":"A Progressive Training Strategy for Vision-Language Models to Counteract Spatio-Temporal Hallucinations in Embodied Reasoning","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2604.09037","citing_title":"SiMing-Bench: Evaluating Procedural Correctness from Continuous Interactions in Clinical Skill Videos","ref_index":37,"is_internal_anchor":true},{"citing_arxiv_id":"2604.07296","citing_title":"OpenSpatial: A Principled Data Engine for Empowering Spatial Intelligence","ref_index":55,"is_internal_anchor":true},{"citing_arxiv_id":"2604.05695","citing_title":"Let Geometry GUIDE: Layer-wise Unrolling of Geometric Priors in Multimodal LLMs","ref_index":47,"is_internal_anchor":true},{"citing_arxiv_id":"2604.18484","citing_title":"XEmbodied: A Foundation Model with Enhanced Geometric and Physical Cues for Large-Scale Embodied Environments","ref_index":107,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04128","citing_title":"JoyAI-Image: Awaking Spatial Intelligence in Unified Multimodal Understanding and Generation","ref_index":96,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT","json":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT.json","graph_json":"https://pith.science/api/pith-number/JRN2MVLBNLQNVM4ZOY4BVW4HFT/graph.json","events_json":"https://pith.science/api/pith-number/JRN2MVLBNLQNVM4ZOY4BVW4HFT/events.json","paper":"https://pith.science/paper/JRN2MVLB"},"agent_actions":{"view_html":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT","download_json":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT.json","view_paper":"https://pith.science/paper/JRN2MVLB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.23764&json=true","fetch_graph":"https://pith.science/api/pith-number/JRN2MVLBNLQNVM4ZOY4BVW4HFT/graph.json","fetch_events":"https://pith.science/api/pith-number/JRN2MVLBNLQNVM4ZOY4BVW4HFT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT/action/storage_attestation","attest_author":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT/action/author_attestation","sign_citation":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT/action/citation_signature","submit_replication":"https://pith.science/pith/JRN2MVLBNLQNVM4ZOY4BVW4HFT/action/replication_record"}},"created_at":"2026-05-26T01:02:26.711305+00:00","updated_at":"2026-05-26T01:02:26.711305+00:00"}