{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:L4AILBMHKY7LREN2AOCQ4RTJB3","short_pith_number":"pith:L4AILBMH","schema_version":"1.0","canonical_sha256":"5f00858587563eb891ba03850e46690ed87506fcb8ce88cdb37b4f48ff764871","source":{"kind":"arxiv","id":"2408.01091","version":2},"attestation_state":"computed","paper":{"title":"Dissecting Dissonance: Benchmarking Large Multimodal Models Against Self-Contradictory Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dequan Wang, Jin Gao, Lei Gan, Yixin Ye, Yuankai Li","submitted_at":"2024-08-02T08:11:11Z","abstract_excerpt":"Large multimodal models (LMMs) excel in adhering to human instructions. However, self-contradictory instructions may arise due to the increasing trend of multimodal interaction and context length, which is challenging for language beginners and vulnerable populations. We introduce the Self-Contradictory Instructions benchmark to evaluate the capability of LMMs in recognizing conflicting commands. It comprises 20,000 conflicts, evenly distributed between language and vision paradigms. It is constructed by a novel automatic dataset creation framework, which expedites the process and enables us t"},"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":"2408.01091","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2024-08-02T08:11:11Z","cross_cats_sorted":[],"title_canon_sha256":"35cd24476f891f3fe7fbdf5677946c7eeba0d1c94ebe8237d87883b2fda63acb","abstract_canon_sha256":"f0a0cfa1533adbef8aada462796ad3de30572d76c027d1b76304cd0fea4dc03e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:51:57.455492Z","signature_b64":"zPZ7lJNyJEk5AJ62yRAeVVb5OBa8rWGl1TTYvOZrYi/JIPqgJ7ZJNeD+xfoptHssaNsamFV6AQY1Hrpu2+UnCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f00858587563eb891ba03850e46690ed87506fcb8ce88cdb37b4f48ff764871","last_reissued_at":"2026-07-05T08:51:57.455053Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:51:57.455053Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Dissecting Dissonance: Benchmarking Large Multimodal Models Against Self-Contradictory Instructions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Dequan Wang, Jin Gao, Lei Gan, Yixin Ye, Yuankai Li","submitted_at":"2024-08-02T08:11:11Z","abstract_excerpt":"Large multimodal models (LMMs) excel in adhering to human instructions. However, self-contradictory instructions may arise due to the increasing trend of multimodal interaction and context length, which is challenging for language beginners and vulnerable populations. We introduce the Self-Contradictory Instructions benchmark to evaluate the capability of LMMs in recognizing conflicting commands. It comprises 20,000 conflicts, evenly distributed between language and vision paradigms. It is constructed by a novel automatic dataset creation framework, which expedites the process and enables us t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.01091","kind":"arxiv","version":2},"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/2408.01091/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":"2408.01091","created_at":"2026-07-05T08:51:57.455109+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.01091v2","created_at":"2026-07-05T08:51:57.455109+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.01091","created_at":"2026-07-05T08:51:57.455109+00:00"},{"alias_kind":"pith_short_12","alias_value":"L4AILBMHKY7L","created_at":"2026-07-05T08:51:57.455109+00:00"},{"alias_kind":"pith_short_16","alias_value":"L4AILBMHKY7LREN2","created_at":"2026-07-05T08:51:57.455109+00:00"},{"alias_kind":"pith_short_8","alias_value":"L4AILBMH","created_at":"2026-07-05T08:51:57.455109+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.04017","citing_title":"Can Large Multimodal Models Actively Recognize Faulty Inputs? A Systematic Evaluation Framework of Their Input Scrutiny Ability","ref_index":9,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3","json":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3.json","graph_json":"https://pith.science/api/pith-number/L4AILBMHKY7LREN2AOCQ4RTJB3/graph.json","events_json":"https://pith.science/api/pith-number/L4AILBMHKY7LREN2AOCQ4RTJB3/events.json","paper":"https://pith.science/paper/L4AILBMH"},"agent_actions":{"view_html":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3","download_json":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3.json","view_paper":"https://pith.science/paper/L4AILBMH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.01091&json=true","fetch_graph":"https://pith.science/api/pith-number/L4AILBMHKY7LREN2AOCQ4RTJB3/graph.json","fetch_events":"https://pith.science/api/pith-number/L4AILBMHKY7LREN2AOCQ4RTJB3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3/action/storage_attestation","attest_author":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3/action/author_attestation","sign_citation":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3/action/citation_signature","submit_replication":"https://pith.science/pith/L4AILBMHKY7LREN2AOCQ4RTJB3/action/replication_record"}},"created_at":"2026-07-05T08:51:57.455109+00:00","updated_at":"2026-07-05T08:51:57.455109+00:00"}