{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:R6S5TYPOCPESYRHVNCROZKFC2Y","short_pith_number":"pith:R6S5TYPO","schema_version":"1.0","canonical_sha256":"8fa5d9e1ee13c92c44f568a2eca8a2d603bc74fa004a07492f10892bac26371b","source":{"kind":"arxiv","id":"2404.12966","version":5},"attestation_state":"computed","paper":{"title":"Look Before You Decide: Prompting Active Deduction of MLLMs for Assumptive Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bin Zhu, Jingjing Chen, Na Zhao, Tianwen Qian, Wentao Tian, Yang Jiao, Yian Li, Yu-Gang Jiang","submitted_at":"2024-04-19T15:53:27Z","abstract_excerpt":"Recently, Multimodal Large Language Models (MLLMs) have achieved significant success across multiple disciplines due to their exceptional instruction-following capabilities and extensive world knowledge. However, whether these MLLMs possess human-like compositional reasoning abilities remains an open problem. To unveil their reasoning behaviors, we first curate a \\textbf{M}ultimodal \\textbf{A}ssumptive \\textbf{R}ea\\textbf{s}oning Benchmark (MARS-Bench) in this paper. Interestingly, we find that most prevalent MLLMs can be easily fooled by the introduction of a presupposition into the question,"},"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":"2404.12966","kind":"arxiv","version":5},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-04-19T15:53:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"2b947499a85f27d99ce8f1d4cff38feaa7dbb47d2839b8feadfdce60eb3ae6d9","abstract_canon_sha256":"d0e75b7fc9733610948b27f910d1a2967510dcdf6a0fbb0240c204b666b20e68"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:50:08.269131Z","signature_b64":"zoCgXxF+C7oYOoUgmF6srLHSMABnsLrntBMmG071jQAybI1noWGzQxcBEVQpPtcangTO7P3TpH1ADKJqddOYAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8fa5d9e1ee13c92c44f568a2eca8a2d603bc74fa004a07492f10892bac26371b","last_reissued_at":"2026-07-05T10:50:08.268643Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:50:08.268643Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Look Before You Decide: Prompting Active Deduction of MLLMs for Assumptive Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bin Zhu, Jingjing Chen, Na Zhao, Tianwen Qian, Wentao Tian, Yang Jiao, Yian Li, Yu-Gang Jiang","submitted_at":"2024-04-19T15:53:27Z","abstract_excerpt":"Recently, Multimodal Large Language Models (MLLMs) have achieved significant success across multiple disciplines due to their exceptional instruction-following capabilities and extensive world knowledge. However, whether these MLLMs possess human-like compositional reasoning abilities remains an open problem. To unveil their reasoning behaviors, we first curate a \\textbf{M}ultimodal \\textbf{A}ssumptive \\textbf{R}ea\\textbf{s}oning Benchmark (MARS-Bench) in this paper. Interestingly, we find that most prevalent MLLMs can be easily fooled by the introduction of a presupposition into the question,"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.12966","kind":"arxiv","version":5},"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/2404.12966/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":"2404.12966","created_at":"2026-07-05T10:50:08.268704+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.12966v5","created_at":"2026-07-05T10:50:08.268704+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.12966","created_at":"2026-07-05T10:50:08.268704+00:00"},{"alias_kind":"pith_short_12","alias_value":"R6S5TYPOCPES","created_at":"2026-07-05T10:50:08.268704+00:00"},{"alias_kind":"pith_short_16","alias_value":"R6S5TYPOCPESYRHV","created_at":"2026-07-05T10:50:08.268704+00:00"},{"alias_kind":"pith_short_8","alias_value":"R6S5TYPO","created_at":"2026-07-05T10:50:08.268704+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.25906","citing_title":"OracleAnalyser: Analysing Implicit Semantics of Oracle Bone Scripts through MLLMs with Post-training","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2605.30346","citing_title":"YoCausal: How Far is Video Generation from World Model? A Causality Perspective","ref_index":72,"is_internal_anchor":false},{"citing_arxiv_id":"2606.28266","citing_title":"RSICCLLM: A Multimodal Large Language Model for Remote Sensing Image Change Captioning","ref_index":29,"is_internal_anchor":false},{"citing_arxiv_id":"2605.09079","citing_title":"CauSim: Scaling Causal Reasoning with Increasingly Complex Causal Simulators","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y","json":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y.json","graph_json":"https://pith.science/api/pith-number/R6S5TYPOCPESYRHVNCROZKFC2Y/graph.json","events_json":"https://pith.science/api/pith-number/R6S5TYPOCPESYRHVNCROZKFC2Y/events.json","paper":"https://pith.science/paper/R6S5TYPO"},"agent_actions":{"view_html":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y","download_json":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y.json","view_paper":"https://pith.science/paper/R6S5TYPO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.12966&json=true","fetch_graph":"https://pith.science/api/pith-number/R6S5TYPOCPESYRHVNCROZKFC2Y/graph.json","fetch_events":"https://pith.science/api/pith-number/R6S5TYPOCPESYRHVNCROZKFC2Y/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y/action/timestamp_anchor","attest_storage":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y/action/storage_attestation","attest_author":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y/action/author_attestation","sign_citation":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y/action/citation_signature","submit_replication":"https://pith.science/pith/R6S5TYPOCPESYRHVNCROZKFC2Y/action/replication_record"}},"created_at":"2026-07-05T10:50:08.268704+00:00","updated_at":"2026-07-05T10:50:08.268704+00:00"}