{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:RER5YUPYNHSU4XUTQTDMJSLXSO","short_pith_number":"pith:RER5YUPY","schema_version":"1.0","canonical_sha256":"8923dc51f869e54e5e9384c6c4c97793ae88204902b6f05716b8b364f8fe8fdb","source":{"kind":"arxiv","id":"2608.04244","version":1},"attestation_state":"computed","paper":{"title":"SIGNPOST-Bench: Benchmarking Text-Vision Conflict Resolution in Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Fan Zhang, Haoxin Lyu, Junting Zhou, Ling Dai, Minghao Liu, Sirun Li, Yong Li","submitted_at":"2026-08-04T21:55:47Z","abstract_excerpt":"Multimodal large language models (MLLMs) make grounded predictions in real-world scenes by combining visual and textual cues, yet existing benchmarks rarely reveal how they arbitrate between these evidence sources when they conflict. We introduce SIGNPOST-Bench, a controlled counterfactual benchmark for evaluating text-vision conflict resolution. Each source image is transformed into a counterfactual quintuplet of Original, Blank, Similar, Random, and Adversarial variants. Synthetic, localized scene-text interventions are designed to preserve non-textual content, enabling paired measurements o"},"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":"2608.04244","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-08-04T21:55:47Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"b41969525d26d406001c5f24f9e4988303ca8b351bb6b5b7b44b3bb8785e3f56","abstract_canon_sha256":"455a18b3788476dcce22ed09ae3e4da4a112af857d8a7f33301b5fd8aca3ef6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-06T01:31:44.365381Z","signature_b64":"Xui/YMdL1bMhw3Ezs14TwQhsRHoRv5sccFAFMyYnBu3SX/Yj8QsNohwF/874e6jS2hyCWd71RrBhTiHSD6wdAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8923dc51f869e54e5e9384c6c4c97793ae88204902b6f05716b8b364f8fe8fdb","last_reissued_at":"2026-08-06T01:31:44.363613Z","signature_status":"signed_v1","first_computed_at":"2026-08-06T01:31:44.363613Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SIGNPOST-Bench: Benchmarking Text-Vision Conflict Resolution in Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.CV","authors_text":"Fan Zhang, Haoxin Lyu, Junting Zhou, Ling Dai, Minghao Liu, Sirun Li, Yong Li","submitted_at":"2026-08-04T21:55:47Z","abstract_excerpt":"Multimodal large language models (MLLMs) make grounded predictions in real-world scenes by combining visual and textual cues, yet existing benchmarks rarely reveal how they arbitrate between these evidence sources when they conflict. We introduce SIGNPOST-Bench, a controlled counterfactual benchmark for evaluating text-vision conflict resolution. Each source image is transformed into a counterfactual quintuplet of Original, Blank, Similar, Random, and Adversarial variants. Synthetic, localized scene-text interventions are designed to preserve non-textual content, enabling paired measurements o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04244","kind":"arxiv","version":1},"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/2608.04244/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":"2608.04244","created_at":"2026-08-06T01:31:44.365676+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.04244v1","created_at":"2026-08-06T01:31:44.365676+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04244","created_at":"2026-08-06T01:31:44.365676+00:00"},{"alias_kind":"pith_short_12","alias_value":"RER5YUPYNHSU","created_at":"2026-08-06T01:31:44.365676+00:00"},{"alias_kind":"pith_short_16","alias_value":"RER5YUPYNHSU4XUT","created_at":"2026-08-06T01:31:44.365676+00:00"},{"alias_kind":"pith_short_8","alias_value":"RER5YUPY","created_at":"2026-08-06T01:31:44.365676+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/RER5YUPYNHSU4XUTQTDMJSLXSO","json":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO.json","graph_json":"https://pith.science/api/pith-number/RER5YUPYNHSU4XUTQTDMJSLXSO/graph.json","events_json":"https://pith.science/api/pith-number/RER5YUPYNHSU4XUTQTDMJSLXSO/events.json","paper":"https://pith.science/paper/RER5YUPY"},"agent_actions":{"view_html":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO","download_json":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO.json","view_paper":"https://pith.science/paper/RER5YUPY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.04244&json=true","fetch_graph":"https://pith.science/api/pith-number/RER5YUPYNHSU4XUTQTDMJSLXSO/graph.json","fetch_events":"https://pith.science/api/pith-number/RER5YUPYNHSU4XUTQTDMJSLXSO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO/action/storage_attestation","attest_author":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO/action/author_attestation","sign_citation":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO/action/citation_signature","submit_replication":"https://pith.science/pith/RER5YUPYNHSU4XUTQTDMJSLXSO/action/replication_record"}},"created_at":"2026-08-06T01:31:44.365676+00:00","updated_at":"2026-08-06T01:31:44.365676+00:00"}