{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:F4QJDZE4R3LWSUF47AGLVDY2NK","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"3f75d75e65730a67bfec2efe00a14adb8290018b7760379439a1a6be189390ef","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-26T16:24:29Z","title_canon_sha256":"ba1d0aa120631af5670471841f99b3b33a324457d23d450b1ce4e350e101a2a5"},"schema_version":"1.0","source":{"id":"2605.27243","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.27243","created_at":"2026-05-27T02:06:07Z"},{"alias_kind":"arxiv_version","alias_value":"2605.27243v1","created_at":"2026-05-27T02:06:07Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27243","created_at":"2026-05-27T02:06:07Z"},{"alias_kind":"pith_short_12","alias_value":"F4QJDZE4R3LW","created_at":"2026-05-27T02:06:07Z"},{"alias_kind":"pith_short_16","alias_value":"F4QJDZE4R3LWSUF4","created_at":"2026-05-27T02:06:07Z"},{"alias_kind":"pith_short_8","alias_value":"F4QJDZE4","created_at":"2026-05-27T02:06:07Z"}],"graph_snapshots":[{"event_id":"sha256:a33093651c0c267bfaa5fcccb54af15391868829b0ae015cd34e82acb0c8cc6c","target":"graph","created_at":"2026-05-27T02:06:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2605.27243/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large vision-language models increasingly rely on long-context modeling to reason over documents, hour-level videos, and long-horizon agent trajectories, requiring them to locate relevant evidence across interleaved text and images. Prior work has studied this behavior using retrieval heads in large language models, but its copy-based criterion does not directly apply when evidence appears in images. We introduce a multimodal retrieval head detection method that scores attention from question tokens to textual or visual evidence. With this method, we show that multimodal retrieval heads are sp","authors_text":"Aaron Branson Cigres Li, Ginny Wong, Haobo Li, Haodong Duan, Lishu Luo, Pasquale Minervini, Simon See, Xiyu Ren, Yangqiu Song, Yiming Du, Yu Zhao, Zhaowei Wang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-26T16:24:29Z","title":"Can Retrieval Heads See Images? Multimodal Retrieval Heads in Long-Context Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27243","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:df0c56524df840006daba1d8720ea32ea0f5ec538797ea43003700fbbd8b9d4e","target":"record","created_at":"2026-05-27T02:06:07Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"3f75d75e65730a67bfec2efe00a14adb8290018b7760379439a1a6be189390ef","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-05-26T16:24:29Z","title_canon_sha256":"ba1d0aa120631af5670471841f99b3b33a324457d23d450b1ce4e350e101a2a5"},"schema_version":"1.0","source":{"id":"2605.27243","kind":"arxiv","version":1}},"canonical_sha256":"2f2091e49c8ed76950bcf80cba8f1a6a929e0214c2f3ee72dcb6aae551540f24","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2f2091e49c8ed76950bcf80cba8f1a6a929e0214c2f3ee72dcb6aae551540f24","first_computed_at":"2026-05-27T02:06:07.382497Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-27T02:06:07.382497Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ExmjDLJNWJDzPL+21httcZsevhn3j+H0bRZTlPZ4VNbHei23MBmFxy3wO9IeHGVFfR+N6mKhmGg42+ScfBF5Bw==","signature_status":"signed_v1","signed_at":"2026-05-27T02:06:07.383035Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.27243","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:df0c56524df840006daba1d8720ea32ea0f5ec538797ea43003700fbbd8b9d4e","sha256:a33093651c0c267bfaa5fcccb54af15391868829b0ae015cd34e82acb0c8cc6c"],"state_sha256":"8bfe8f7dbd61557351e2273e893b9802b23be36acea66bc7bd3426a1c51ffa4c"}