{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:WDFRUYQYNRC5DHJMEJVBRRXYPZ","short_pith_number":"pith:WDFRUYQY","schema_version":"1.0","canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","source":{"kind":"arxiv","id":"2605.29585","version":1},"attestation_state":"computed","paper":{"title":"World Models in Words: Auditing Physical State-Transition Commitments in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Emmanuelle Bourigault","submitted_at":"2026-05-28T08:29:32Z","abstract_excerpt":"Vision-language models (VLMs) are increasingly used to answer questions about physical scenes, yet most evaluations reduce performance to a final answer. This hides whether the model perceived the right objects, represented the right physical state, predicted a plausible transition, or merely selected the right option for the wrong reasons. We introduce \\wmw, an evaluation framework for auditing the \\emph{language-expressed physical commitments} of VLMs. Instead of scoring only $I,q\\mapsto a$, we ask models to produce a typed trace $I,q\\mapsto(s_0,\\Delta s,s_1,a)$: an initial state, a state tr"},"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":"2605.29585","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T08:29:32Z","cross_cats_sorted":[],"title_canon_sha256":"c8467f2955c600d8a79383e144070c99f6950351ab9243d677e11bff3596ec72","abstract_canon_sha256":"079a25a1ed246100b9cd8b7211bf73a29dee326bd6a17316bfe2721a6d0700bc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:05:49.107554Z","signature_b64":"F0nj9Hmn6dElJ45gIpZwc/R8zLPrWwndr/lR7BpHS+ZkUaW+yNUKqxHIsGwm0xQRGmmoy4w2b3CZcKuzviImBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","last_reissued_at":"2026-05-29T01:05:49.106829Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:05:49.106829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"World Models in Words: Auditing Physical State-Transition Commitments in Vision-Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Emmanuelle Bourigault","submitted_at":"2026-05-28T08:29:32Z","abstract_excerpt":"Vision-language models (VLMs) are increasingly used to answer questions about physical scenes, yet most evaluations reduce performance to a final answer. This hides whether the model perceived the right objects, represented the right physical state, predicted a plausible transition, or merely selected the right option for the wrong reasons. We introduce \\wmw, an evaluation framework for auditing the \\emph{language-expressed physical commitments} of VLMs. Instead of scoring only $I,q\\mapsto a$, we ask models to produce a typed trace $I,q\\mapsto(s_0,\\Delta s,s_1,a)$: an initial state, a state tr"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29585","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/2605.29585/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":"2605.29585","created_at":"2026-05-29T01:05:49.106949+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.29585v1","created_at":"2026-05-29T01:05:49.106949+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29585","created_at":"2026-05-29T01:05:49.106949+00:00"},{"alias_kind":"pith_short_12","alias_value":"WDFRUYQYNRC5","created_at":"2026-05-29T01:05:49.106949+00:00"},{"alias_kind":"pith_short_16","alias_value":"WDFRUYQYNRC5DHJM","created_at":"2026-05-29T01:05:49.106949+00:00"},{"alias_kind":"pith_short_8","alias_value":"WDFRUYQY","created_at":"2026-05-29T01:05:49.106949+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ","json":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ.json","graph_json":"https://pith.science/api/pith-number/WDFRUYQYNRC5DHJMEJVBRRXYPZ/graph.json","events_json":"https://pith.science/api/pith-number/WDFRUYQYNRC5DHJMEJVBRRXYPZ/events.json","paper":"https://pith.science/paper/WDFRUYQY"},"agent_actions":{"view_html":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ","download_json":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ.json","view_paper":"https://pith.science/paper/WDFRUYQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.29585&json=true","fetch_graph":"https://pith.science/api/pith-number/WDFRUYQYNRC5DHJMEJVBRRXYPZ/graph.json","fetch_events":"https://pith.science/api/pith-number/WDFRUYQYNRC5DHJMEJVBRRXYPZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/action/storage_attestation","attest_author":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/action/author_attestation","sign_citation":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/action/citation_signature","submit_replication":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/action/replication_record"}},"created_at":"2026-05-29T01:05:49.106949+00:00","updated_at":"2026-05-29T01:05:49.106949+00:00"}