{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:YOAX255MLGK57NNVGX3CZ4POSU","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":"b0c9629294426b0a9bcec3b33ffcacda25f386e9ad045ff93db133d02dd82667","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T16:14:44Z","title_canon_sha256":"7d591faafb7b2471706857ee16b2530c1ab4752fc24122f6df97aa42f4045b70"},"schema_version":"1.0","source":{"id":"2605.13737","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.13737","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"arxiv_version","alias_value":"2605.13737v1","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13737","created_at":"2026-05-18T02:44:16Z"},{"alias_kind":"pith_short_12","alias_value":"YOAX255MLGK5","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_16","alias_value":"YOAX255MLGK57NNV","created_at":"2026-05-18T12:33:37Z"},{"alias_kind":"pith_short_8","alias_value":"YOAX255M","created_at":"2026-05-18T12:33:37Z"}],"graph_snapshots":[{"event_id":"sha256:080ffc38fb6b9a490d53eb2a54a705c41f8428512d03bb5edefd60d300e8d3aa","target":"graph","created_at":"2026-05-18T02:44:16Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"hidden states reliably encode premise-perception mismatches even when the same models almost never reject the false claim in their outputs"},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the IMAVB benchmark's curated misleading premises cleanly isolate grounding failures without introducing confounds from clip selection, question phrasing, or model-specific training data, and that probe-based detection of hidden-state mismatches accurately reflects functional encoding."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"Omnimodal LLMs encode premise-perception mismatches in hidden states yet almost never reject false textual claims, exposing a representation-action gap that is modality-asymmetric and prompt-resistant."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs."}],"snapshot_sha256":"00b3ef468fc7f11ebb122098271f2a83121b4db0528edfff7e855bc1b2bccb7d"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"When an omnimodal large language model accepts a question whose textual premise contradicts what it actually sees or hears, does the failure lie in perception or in action? Recent omnimodal models are positioned as perception-grounded agents that jointly process video, audio, and text, yet a basic form of grounding remains untested: catching a textual claim that conflicts with the model's own sensory input. We introduce IMAVB, a curated 500-clip benchmark of long-form movies with a 2x2 design crossing target modality (vision, audio) and premise condition (standard, misleading), which lets us m","authors_text":"Fanyi Pu, Kaichen Zhang, Shuo Sun, Trung Nguyen Quang, Yiming Gao, Ziwei Liu","cross_cats":["cs.CL"],"headline":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs.","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T16:14:44Z","title":"Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs"},"references":{"count":99,"internal_anchors":3,"resolved_work":99,"sample":[{"cited_arxiv_id":"2601.03267","doi":"","is_internal_anchor":true,"ref_index":1,"title":"OpenAI GPT-5 System Card","work_id":"ca87689a-0d29-4476-b504-b65dbbb08af4","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"A new era of intelligence with Gemini 3, 2026","work_id":"70678dd4-c18e-407f-a968-a0b8180cf72a","year":2026},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Qwen3-omni technical report, 2025","work_id":"2421f421-da09-44ef-b96f-2a64fba6a7ef","year":2025},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"Amos Azaria and Tom M. Mitchell. The internal state of an llm knows when it’s lying, 2023","work_id":"ed116ea6-70d9-4c0f-90db-16fe14dcc3a7","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":5,"title":"Discovering latent knowledge in language models without supervision","work_id":"db51027f-63a1-4026-b9c6-6df48de0cf76","year":2022}],"snapshot_sha256":"a67f6b2d0dce6c7852eba2d59c852fa2e4a799e1074b11a14dc3994b61ca4a2f"},"source":{"id":"2605.13737","kind":"arxiv","version":1},"verdict":{"created_at":"2026-05-14T18:03:22.061288Z","id":"009deb09-86c0-4403-82bf-ac51277f9fda","model_set":{"reader":"grok-4.3"},"one_line_summary":"Omnimodal LLMs encode premise-perception mismatches in hidden states yet almost never reject false textual claims, exposing a representation-action gap that is modality-asymmetric and prompt-resistant.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs.","strongest_claim":"hidden states reliably encode premise-perception mismatches even when the same models almost never reject the false claim in their outputs","weakest_assumption":"That the IMAVB benchmark's curated misleading premises cleanly isolate grounding failures without introducing confounds from clip selection, question phrasing, or model-specific training data, and that probe-based detection of hidden-state mismatches accurately reflects functional encoding."}},"verdict_id":"009deb09-86c0-4403-82bf-ac51277f9fda"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:836e7dfdf89c6bcffc50daad25b2a86db900cd61e8d3dfda619b60bc49f28290","target":"record","created_at":"2026-05-18T02:44:16Z","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":"b0c9629294426b0a9bcec3b33ffcacda25f386e9ad045ff93db133d02dd82667","cross_cats_sorted":["cs.CL"],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T16:14:44Z","title_canon_sha256":"7d591faafb7b2471706857ee16b2530c1ab4752fc24122f6df97aa42f4045b70"},"schema_version":"1.0","source":{"id":"2605.13737","kind":"arxiv","version":1}},"canonical_sha256":"c3817d77ac5995dfb5b535f62cf1ee9511d405cbfe4962a03d36759b9c4f1026","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c3817d77ac5995dfb5b535f62cf1ee9511d405cbfe4962a03d36759b9c4f1026","first_computed_at":"2026-05-18T02:44:16.505386Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T02:44:16.505386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"oKf+a1WO+ynPj+FYh2If7hUaoydk2LyMPvIf2WmVoDPU2FjR3b/pfGP5fL7f8aSWIxgxolMHEGVteE1twjLfBQ==","signature_status":"signed_v1","signed_at":"2026-05-18T02:44:16.505843Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.13737","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:836e7dfdf89c6bcffc50daad25b2a86db900cd61e8d3dfda619b60bc49f28290","sha256:080ffc38fb6b9a490d53eb2a54a705c41f8428512d03bb5edefd60d300e8d3aa"],"state_sha256":"b88948e2372ce12a7fc68264f92c7e6adad41fb6c68d102f927d1bfa934f91d2"}