{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:IMI3GSBIIICJAQI6MR6EX5U37F","short_pith_number":"pith:IMI3GSBI","schema_version":"1.0","canonical_sha256":"4311b34828420490411e647c4bf69bf9794d26311711df6a5b84371781209fb5","source":{"kind":"arxiv","id":"2505.24238","version":2},"attestation_state":"computed","paper":{"title":"MIRAGE: Assessing Hallucination in Multimodal Reasoning Chains of MLLM","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bowen Dong, Guanglei Yang, Lei Zhang, Minheng Ni, Wangmeng Zuo, Zitong Huang","submitted_at":"2025-05-30T05:54:36Z","abstract_excerpt":"Multimodal hallucination in multimodal large language models (MLLMs) restricts the correctness of MLLMs. However, multimodal hallucinations are multi-sourced and arise from diverse causes. Existing benchmarks fail to adequately distinguish between perception-induced hallucinations and reasoning-induced hallucinations. This failure constitutes a significant issue and hinders the diagnosis of multimodal reasoning failures within MLLMs. To address this, we propose the {\\dataset} benchmark, which isolates reasoning hallucinations by constructing questions where input images are correctly perceived"},"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":"2505.24238","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-05-30T05:54:36Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"4c9bbf96de93778298b022727d631a5fe6bc2f9ffa8dc6f16a591b19cccdbb79","abstract_canon_sha256":"0585d689d5a9a51d5041dfe04f515120719c5253bd516a7e9d6e0043ca85d017"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:14:09.158667Z","signature_b64":"AEJqREXxpqgn0y0yqltYNRt/+H1LMsL9lmBt5SBUqzY46/Z/WqvBFzVv4fIqsxaL3/W/B6NduKIMUNVlSniSDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4311b34828420490411e647c4bf69bf9794d26311711df6a5b84371781209fb5","last_reissued_at":"2026-07-05T11:14:09.158156Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:14:09.158156Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MIRAGE: Assessing Hallucination in Multimodal Reasoning Chains of MLLM","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Bowen Dong, Guanglei Yang, Lei Zhang, Minheng Ni, Wangmeng Zuo, Zitong Huang","submitted_at":"2025-05-30T05:54:36Z","abstract_excerpt":"Multimodal hallucination in multimodal large language models (MLLMs) restricts the correctness of MLLMs. However, multimodal hallucinations are multi-sourced and arise from diverse causes. Existing benchmarks fail to adequately distinguish between perception-induced hallucinations and reasoning-induced hallucinations. This failure constitutes a significant issue and hinders the diagnosis of multimodal reasoning failures within MLLMs. To address this, we propose the {\\dataset} benchmark, which isolates reasoning hallucinations by constructing questions where input images are correctly perceived"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.24238","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/2505.24238/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":"2505.24238","created_at":"2026-07-05T11:14:09.158216+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.24238v2","created_at":"2026-07-05T11:14:09.158216+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.24238","created_at":"2026-07-05T11:14:09.158216+00:00"},{"alias_kind":"pith_short_12","alias_value":"IMI3GSBIIICJ","created_at":"2026-07-05T11:14:09.158216+00:00"},{"alias_kind":"pith_short_16","alias_value":"IMI3GSBIIICJAQI6","created_at":"2026-07-05T11:14:09.158216+00:00"},{"alias_kind":"pith_short_8","alias_value":"IMI3GSBI","created_at":"2026-07-05T11:14:09.158216+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.03179","citing_title":"Understanding the Role of Hallucination in Reinforcement Post-Training of Multimodal Reasoning Models","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.23392","citing_title":"SoccerRef-Agents: Multi-Agent System for Automated Soccer Refereeing","ref_index":5,"is_internal_anchor":false},{"citing_arxiv_id":"2604.21027","citing_title":"HypEHR: Hyperbolic Modeling of Electronic Health Records for Efficient Question Answering","ref_index":230,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F","json":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F.json","graph_json":"https://pith.science/api/pith-number/IMI3GSBIIICJAQI6MR6EX5U37F/graph.json","events_json":"https://pith.science/api/pith-number/IMI3GSBIIICJAQI6MR6EX5U37F/events.json","paper":"https://pith.science/paper/IMI3GSBI"},"agent_actions":{"view_html":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F","download_json":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F.json","view_paper":"https://pith.science/paper/IMI3GSBI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.24238&json=true","fetch_graph":"https://pith.science/api/pith-number/IMI3GSBIIICJAQI6MR6EX5U37F/graph.json","fetch_events":"https://pith.science/api/pith-number/IMI3GSBIIICJAQI6MR6EX5U37F/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F/action/storage_attestation","attest_author":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F/action/author_attestation","sign_citation":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F/action/citation_signature","submit_replication":"https://pith.science/pith/IMI3GSBIIICJAQI6MR6EX5U37F/action/replication_record"}},"created_at":"2026-07-05T11:14:09.158216+00:00","updated_at":"2026-07-05T11:14:09.158216+00:00"}