{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:YW4VTQRLRG2EE5K76VRM5WKWW4","short_pith_number":"pith:YW4VTQRL","schema_version":"1.0","canonical_sha256":"c5b959c22b89b442755ff562ced956b71734660b880d2e35125ec83f6edd532d","source":{"kind":"arxiv","id":"2508.05405","version":1},"attestation_state":"computed","paper":{"title":"DeepPHY: Benchmarking Agentic VLMs on Physical Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"B\\\"orje F. Karlsson, Bo Zheng, Jun Song, Pi Bu, Qi Zhu, Tengtao Song, Xinrun Xu, Ye Wang, Zhiming Ding, Ziming Wang","submitted_at":"2025-08-07T13:58:19Z","abstract_excerpt":"Although Vision Language Models (VLMs) exhibit strong perceptual abilities and impressive visual reasoning, they struggle with attention to detail and precise action planning in complex, dynamic environments, leading to subpar performance. Real-world tasks typically require complex interactions, advanced spatial reasoning, long-term planning, and continuous strategy refinement, usually necessitating understanding the physics rules of the target scenario. However, evaluating these capabilities in real-world scenarios is often prohibitively expensive. To bridge this gap, we introduce DeepPHY, a "},"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":"2508.05405","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2025-08-07T13:58:19Z","cross_cats_sorted":[],"title_canon_sha256":"3914e5881a158780ab33129bd397954b4095061ddf315d5c2527a669c7810da6","abstract_canon_sha256":"73d8b3b7f17f439c235dcef853d73f376424e3d0cf0a939adabf40dd21465fba"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:50:12.575911Z","signature_b64":"2/BcVQGQaVHclvpzaPMi3/xL/Ga8gppDs5jPqKTIF9vrVOMQ3njcEhFgnNarbjLrvOU8HhbxiJ02/qlgS5QXAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c5b959c22b89b442755ff562ced956b71734660b880d2e35125ec83f6edd532d","last_reissued_at":"2026-07-05T11:50:12.575426Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:50:12.575426Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepPHY: Benchmarking Agentic VLMs on Physical Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"B\\\"orje F. Karlsson, Bo Zheng, Jun Song, Pi Bu, Qi Zhu, Tengtao Song, Xinrun Xu, Ye Wang, Zhiming Ding, Ziming Wang","submitted_at":"2025-08-07T13:58:19Z","abstract_excerpt":"Although Vision Language Models (VLMs) exhibit strong perceptual abilities and impressive visual reasoning, they struggle with attention to detail and precise action planning in complex, dynamic environments, leading to subpar performance. Real-world tasks typically require complex interactions, advanced spatial reasoning, long-term planning, and continuous strategy refinement, usually necessitating understanding the physics rules of the target scenario. However, evaluating these capabilities in real-world scenarios is often prohibitively expensive. To bridge this gap, we introduce DeepPHY, a "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.05405","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/2508.05405/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":"2508.05405","created_at":"2026-07-05T11:50:12.575483+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.05405v1","created_at":"2026-07-05T11:50:12.575483+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.05405","created_at":"2026-07-05T11:50:12.575483+00:00"},{"alias_kind":"pith_short_12","alias_value":"YW4VTQRLRG2E","created_at":"2026-07-05T11:50:12.575483+00:00"},{"alias_kind":"pith_short_16","alias_value":"YW4VTQRLRG2EE5K7","created_at":"2026-07-05T11:50:12.575483+00:00"},{"alias_kind":"pith_short_8","alias_value":"YW4VTQRL","created_at":"2026-07-05T11:50:12.575483+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.06522","citing_title":"Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2512.23292","citing_title":"Agentic Physical AI toward a Domain-Specific Foundation Model for Nuclear Reactor Control","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4","json":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4.json","graph_json":"https://pith.science/api/pith-number/YW4VTQRLRG2EE5K76VRM5WKWW4/graph.json","events_json":"https://pith.science/api/pith-number/YW4VTQRLRG2EE5K76VRM5WKWW4/events.json","paper":"https://pith.science/paper/YW4VTQRL"},"agent_actions":{"view_html":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4","download_json":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4.json","view_paper":"https://pith.science/paper/YW4VTQRL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.05405&json=true","fetch_graph":"https://pith.science/api/pith-number/YW4VTQRLRG2EE5K76VRM5WKWW4/graph.json","fetch_events":"https://pith.science/api/pith-number/YW4VTQRLRG2EE5K76VRM5WKWW4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4/action/storage_attestation","attest_author":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4/action/author_attestation","sign_citation":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4/action/citation_signature","submit_replication":"https://pith.science/pith/YW4VTQRLRG2EE5K76VRM5WKWW4/action/replication_record"}},"created_at":"2026-07-05T11:50:12.575483+00:00","updated_at":"2026-07-05T11:50:12.575483+00:00"}