{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:WDFRUYQYNRC5DHJMEJVBRRXYPZ","short_pith_number":"pith:WDFRUYQY","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"},"canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","source":{"kind":"arxiv","id":"2605.29585","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.29585","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.29585v1","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29585","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_12","alias_value":"WDFRUYQYNRC5","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_16","alias_value":"WDFRUYQYNRC5DHJM","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_8","alias_value":"WDFRUYQY","created_at":"2026-05-29T01:05:49Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:WDFRUYQYNRC5DHJMEJVBRRXYPZ","target":"record","payload":{"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"},"canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","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"},"source_kind":"arxiv","source_id":"2605.29585","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T01:05:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Y+A3nHa6aipaHtqpb42I8o8fvhB5VTk0h53SwNGya0p/GP1vvS0ZkdV+Bhhn4cmhi4POj7VlDq8WGuzveYakAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T05:53:17.107827Z"},"content_sha256":"69b91519dd78a8151ea5e39e862c71ff765d160e2b5ac34f8c5e36dc8a305296","schema_version":"1.0","event_id":"sha256:69b91519dd78a8151ea5e39e862c71ff765d160e2b5ac34f8c5e36dc8a305296"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:WDFRUYQYNRC5DHJMEJVBRRXYPZ","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-29T01:05:49Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"XaQ0fC6ymeKxxv2ThHsuA6nE4G1vuZhrUdrVKX2v80Hwi8eAvveCBgrmEG5SaOjGZ89b8ZANUVTBcCGNw0VVCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T05:53:17.108204Z"},"content_sha256":"3cb76f88da35dac67b8c288b8174f0b952302e06f27574165c8fb6bcce49f7bd","schema_version":"1.0","event_id":"sha256:3cb76f88da35dac67b8c288b8174f0b952302e06f27574165c8fb6bcce49f7bd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/bundle.json","state_url":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-30T05:53:17Z","links":{"resolver":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ","bundle":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/bundle.json","state":"https://pith.science/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WDFRUYQYNRC5DHJMEJVBRRXYPZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:WDFRUYQYNRC5DHJMEJVBRRXYPZ","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"079a25a1ed246100b9cd8b7211bf73a29dee326bd6a17316bfe2721a6d0700bc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T08:29:32Z","title_canon_sha256":"c8467f2955c600d8a79383e144070c99f6950351ab9243d677e11bff3596ec72"},"schema_version":"1.0","source":{"id":"2605.29585","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.29585","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"arxiv_version","alias_value":"2605.29585v1","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.29585","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_12","alias_value":"WDFRUYQYNRC5","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_16","alias_value":"WDFRUYQYNRC5DHJM","created_at":"2026-05-29T01:05:49Z"},{"alias_kind":"pith_short_8","alias_value":"WDFRUYQY","created_at":"2026-05-29T01:05:49Z"}],"graph_snapshots":[{"event_id":"sha256:3cb76f88da35dac67b8c288b8174f0b952302e06f27574165c8fb6bcce49f7bd","target":"graph","created_at":"2026-05-29T01:05:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.29585/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Emmanuelle Bourigault","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T08:29:32Z","title":"World Models in Words: Auditing Physical State-Transition Commitments in Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.29585","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:69b91519dd78a8151ea5e39e862c71ff765d160e2b5ac34f8c5e36dc8a305296","target":"record","created_at":"2026-05-29T01:05:49Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"079a25a1ed246100b9cd8b7211bf73a29dee326bd6a17316bfe2721a6d0700bc","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-28T08:29:32Z","title_canon_sha256":"c8467f2955c600d8a79383e144070c99f6950351ab9243d677e11bff3596ec72"},"schema_version":"1.0","source":{"id":"2605.29585","kind":"arxiv","version":1}},"canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b0cb1a62186c45d19d2c226a18c6f87e66ba2cc0bba698025e8f0e35f591e238","first_computed_at":"2026-05-29T01:05:49.106829Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-29T01:05:49.106829Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"F0nj9Hmn6dElJ45gIpZwc/R8zLPrWwndr/lR7BpHS+ZkUaW+yNUKqxHIsGwm0xQRGmmoy4w2b3CZcKuzviImBA==","signature_status":"signed_v1","signed_at":"2026-05-29T01:05:49.107554Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.29585","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:69b91519dd78a8151ea5e39e862c71ff765d160e2b5ac34f8c5e36dc8a305296","sha256:3cb76f88da35dac67b8c288b8174f0b952302e06f27574165c8fb6bcce49f7bd"],"state_sha256":"ebc82ff26a26f0c636a0f18e2e023f183c53ab828c5b22c6f80d1b295749ce3f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VzEaAhObSWVnzRtcJlxCEXd4XQl3EbbvH1pix2leTLYjekUiSjtTtt2b4hO9FCs0DmnJzn58W9Yre/aJdfXeCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T05:53:17.110664Z","bundle_sha256":"20ecbf292bbd580d17c67550c3d9fbc0757eb0994a78f5e74009a00b36d9158b"}}