{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:YOAX255MLGK57NNVGX3CZ4POSU","short_pith_number":"pith:YOAX255M","schema_version":"1.0","canonical_sha256":"c3817d77ac5995dfb5b535f62cf1ee9511d405cbfe4962a03d36759b9c4f1026","source":{"kind":"arxiv","id":"2605.13737","version":1},"attestation_state":"computed","paper":{"title":"Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Fanyi Pu, Kaichen Zhang, Shuo Sun, Trung Nguyen Quang, Yiming Gao, Ziwei Liu","submitted_at":"2026-05-13T16:14:44Z","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"},"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":true,"formal_links_present":false},"canonical_record":{"source":{"id":"2605.13737","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-05-13T16:14:44Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"7d591faafb7b2471706857ee16b2530c1ab4752fc24122f6df97aa42f4045b70","abstract_canon_sha256":"b0c9629294426b0a9bcec3b33ffcacda25f386e9ad045ff93db133d02dd82667"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T02:44:16.505843Z","signature_b64":"oKf+a1WO+ynPj+FYh2If7hUaoydk2LyMPvIf2WmVoDPU2FjR3b/pfGP5fL7f8aSWIxgxolMHEGVteE1twjLfBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c3817d77ac5995dfb5b535f62cf1ee9511d405cbfe4962a03d36759b9c4f1026","last_reissued_at":"2026-05-18T02:44:16.505386Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T02:44:16.505386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Senses Wide Shut: A Representation-Action Gap in Omnimodal LLMs","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs.","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Fanyi Pu, Kaichen Zhang, Shuo Sun, Trung Nguyen Quang, Yiming Gao, Ziwei Liu","submitted_at":"2026-05-13T16:14:44Z","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"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"hidden states reliably encode premise-perception mismatches even when the same models almost never reject the false claim in their outputs","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","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.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","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.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"00b3ef468fc7f11ebb122098271f2a83121b4db0528edfff7e855bc1b2bccb7d"},"source":{"id":"2605.13737","kind":"arxiv","version":1},"verdict":{"id":"009deb09-86c0-4403-82bf-ac51277f9fda","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T18:03:22.061288Z","strongest_claim":"hidden states reliably encode premise-perception mismatches even when the same models almost never reject the false claim in their outputs","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","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.","pith_extraction_headline":"Omnimodal LLMs encode premise-perception mismatches in hidden states but almost never reject the conflicting claims in their outputs."},"references":{"count":99,"sample":[{"doi":"","year":2026,"title":"OpenAI GPT-5 System Card","work_id":"ca87689a-0d29-4476-b504-b65dbbb08af4","ref_index":1,"cited_arxiv_id":"2601.03267","is_internal_anchor":true},{"doi":"","year":2026,"title":"A new era of intelligence with Gemini 3, 2026","work_id":"70678dd4-c18e-407f-a968-a0b8180cf72a","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2025,"title":"Qwen3-omni technical report, 2025","work_id":"2421f421-da09-44ef-b96f-2a64fba6a7ef","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"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","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Discovering latent knowledge in language models without supervision","work_id":"db51027f-63a1-4026-b9c6-6df48de0cf76","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":99,"snapshot_sha256":"a67f6b2d0dce6c7852eba2d59c852fa2e4a799e1074b11a14dc3994b61ca4a2f","internal_anchors":3},"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":"2605.13737","created_at":"2026-05-18T02:44:16.505462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.13737v1","created_at":"2026-05-18T02:44:16.505462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.13737","created_at":"2026-05-18T02:44:16.505462+00:00"},{"alias_kind":"pith_short_12","alias_value":"YOAX255MLGK5","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_16","alias_value":"YOAX255MLGK57NNV","created_at":"2026-05-18T12:33:37.589309+00:00"},{"alias_kind":"pith_short_8","alias_value":"YOAX255M","created_at":"2026-05-18T12:33:37.589309+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/YOAX255MLGK57NNVGX3CZ4POSU","json":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU.json","graph_json":"https://pith.science/api/pith-number/YOAX255MLGK57NNVGX3CZ4POSU/graph.json","events_json":"https://pith.science/api/pith-number/YOAX255MLGK57NNVGX3CZ4POSU/events.json","paper":"https://pith.science/paper/YOAX255M"},"agent_actions":{"view_html":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU","download_json":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU.json","view_paper":"https://pith.science/paper/YOAX255M","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.13737&json=true","fetch_graph":"https://pith.science/api/pith-number/YOAX255MLGK57NNVGX3CZ4POSU/graph.json","fetch_events":"https://pith.science/api/pith-number/YOAX255MLGK57NNVGX3CZ4POSU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU/action/storage_attestation","attest_author":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU/action/author_attestation","sign_citation":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU/action/citation_signature","submit_replication":"https://pith.science/pith/YOAX255MLGK57NNVGX3CZ4POSU/action/replication_record"}},"created_at":"2026-05-18T02:44:16.505462+00:00","updated_at":"2026-05-18T02:44:16.505462+00:00"}