{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:HFLLX4QDAWWUNZBEZNDIQQT46K","short_pith_number":"pith:HFLLX4QD","canonical_record":{"source":{"id":"2606.30168","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T11:44:09Z","cross_cats_sorted":[],"title_canon_sha256":"a8b91d19810ed29a7a07a9023a6ee5e37293cf510a7656cb3e218aabdf17b6a3","abstract_canon_sha256":"84d925be7ad73514c5b0082ff785a9d96ac6535c4f8ffc71b97b394d8a70cf93"},"schema_version":"1.0"},"canonical_sha256":"3956bbf20305ad46e424cb4688427cf2b2bba96aab71190b9e47fa2efcdc0730","source":{"kind":"arxiv","id":"2606.30168","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.30168","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"arxiv_version","alias_value":"2606.30168v1","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30168","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_12","alias_value":"HFLLX4QDAWWU","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_16","alias_value":"HFLLX4QDAWWUNZBE","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_8","alias_value":"HFLLX4QD","created_at":"2026-06-30T02:17:52Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:HFLLX4QDAWWUNZBEZNDIQQT46K","target":"record","payload":{"canonical_record":{"source":{"id":"2606.30168","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T11:44:09Z","cross_cats_sorted":[],"title_canon_sha256":"a8b91d19810ed29a7a07a9023a6ee5e37293cf510a7656cb3e218aabdf17b6a3","abstract_canon_sha256":"84d925be7ad73514c5b0082ff785a9d96ac6535c4f8ffc71b97b394d8a70cf93"},"schema_version":"1.0"},"canonical_sha256":"3956bbf20305ad46e424cb4688427cf2b2bba96aab71190b9e47fa2efcdc0730","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:52.358176Z","signature_b64":"yJ2i5jlAy+LRqLEAII7UELomnz219YKefFNUhMqmmytO+phjDcnOrczUa779+/6RxTrvzCjyAHU/GRhx3GCbDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3956bbf20305ad46e424cb4688427cf2b2bba96aab71190b9e47fa2efcdc0730","last_reissued_at":"2026-06-30T02:17:52.357630Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:52.357630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2606.30168","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-30T02:17:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"njWXcO/ymSs+cQFCezAHCfOJcEsrtfuDBmsul2KQyGjWHnEZhXBlH4LiCVRytl4gJDn3NnH0Un27uaRAdPaGBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T17:27:05.916464Z"},"content_sha256":"dfa0d0e0c232b344b443cefce8d65d7bd986fc0cab1c52dba45e508aac146b5a","schema_version":"1.0","event_id":"sha256:dfa0d0e0c232b344b443cefce8d65d7bd986fc0cab1c52dba45e508aac146b5a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:HFLLX4QDAWWUNZBEZNDIQQT46K","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Latent Noise Mask for Reducing Visual Redundancy in Multimodal Large Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Hongyuan Zhang, Kai Jiang, Ruishu Zhu, Siqi Huang, Xuelong Li","submitted_at":"2026-06-29T11:44:09Z","abstract_excerpt":"Multimodal large language models (MLLMs) often fail in fine-grained visual reasoning, as question-relevant visual cues are diluted by dense and redundant image tokens. Recent multimodal reasoning methods usually extend chain-of-thought from language models into visual or latent spaces, seeking to add intermediate reasoning states while overlooking the negative impact of redundant visual tokens. We propose LatEnt Noise maSk (Lens), a question-conditioned visual evidence purification framework that empowers MLLMs to reason with cleaner visual cues in latent space. Lens introduces a lightweight L"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30168","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/2606.30168/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-30T02:17:52Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"caoic5FM7+n+QkdC0FPc0uBBKVAd6l/PzbRpJuKlCRsaJ2ksVLmA+oqzvJUps2XGEyMjWgumhyBJm7FYAthaDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-30T17:27:05.916836Z"},"content_sha256":"1c15d6f31e2334c089da46b44939883879679608d006352547cb17a2b801d467","schema_version":"1.0","event_id":"sha256:1c15d6f31e2334c089da46b44939883879679608d006352547cb17a2b801d467"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/bundle.json","state_url":"https://pith.science/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-30T17:27:05Z","links":{"resolver":"https://pith.science/pith/HFLLX4QDAWWUNZBEZNDIQQT46K","bundle":"https://pith.science/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/bundle.json","state":"https://pith.science/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HFLLX4QDAWWUNZBEZNDIQQT46K/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:HFLLX4QDAWWUNZBEZNDIQQT46K","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":"84d925be7ad73514c5b0082ff785a9d96ac6535c4f8ffc71b97b394d8a70cf93","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T11:44:09Z","title_canon_sha256":"a8b91d19810ed29a7a07a9023a6ee5e37293cf510a7656cb3e218aabdf17b6a3"},"schema_version":"1.0","source":{"id":"2606.30168","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.30168","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"arxiv_version","alias_value":"2606.30168v1","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.30168","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_12","alias_value":"HFLLX4QDAWWU","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_16","alias_value":"HFLLX4QDAWWUNZBE","created_at":"2026-06-30T02:17:52Z"},{"alias_kind":"pith_short_8","alias_value":"HFLLX4QD","created_at":"2026-06-30T02:17:52Z"}],"graph_snapshots":[{"event_id":"sha256:1c15d6f31e2334c089da46b44939883879679608d006352547cb17a2b801d467","target":"graph","created_at":"2026-06-30T02:17:52Z","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/2606.30168/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Multimodal large language models (MLLMs) often fail in fine-grained visual reasoning, as question-relevant visual cues are diluted by dense and redundant image tokens. Recent multimodal reasoning methods usually extend chain-of-thought from language models into visual or latent spaces, seeking to add intermediate reasoning states while overlooking the negative impact of redundant visual tokens. We propose LatEnt Noise maSk (Lens), a question-conditioned visual evidence purification framework that empowers MLLMs to reason with cleaner visual cues in latent space. Lens introduces a lightweight L","authors_text":"Hongyuan Zhang, Kai Jiang, Ruishu Zhu, Siqi Huang, Xuelong Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T11:44:09Z","title":"Latent Noise Mask for Reducing Visual Redundancy in Multimodal Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.30168","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:dfa0d0e0c232b344b443cefce8d65d7bd986fc0cab1c52dba45e508aac146b5a","target":"record","created_at":"2026-06-30T02:17:52Z","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":"84d925be7ad73514c5b0082ff785a9d96ac6535c4f8ffc71b97b394d8a70cf93","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T11:44:09Z","title_canon_sha256":"a8b91d19810ed29a7a07a9023a6ee5e37293cf510a7656cb3e218aabdf17b6a3"},"schema_version":"1.0","source":{"id":"2606.30168","kind":"arxiv","version":1}},"canonical_sha256":"3956bbf20305ad46e424cb4688427cf2b2bba96aab71190b9e47fa2efcdc0730","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3956bbf20305ad46e424cb4688427cf2b2bba96aab71190b9e47fa2efcdc0730","first_computed_at":"2026-06-30T02:17:52.357630Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-30T02:17:52.357630Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"yJ2i5jlAy+LRqLEAII7UELomnz219YKefFNUhMqmmytO+phjDcnOrczUa779+/6RxTrvzCjyAHU/GRhx3GCbDg==","signature_status":"signed_v1","signed_at":"2026-06-30T02:17:52.358176Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.30168","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dfa0d0e0c232b344b443cefce8d65d7bd986fc0cab1c52dba45e508aac146b5a","sha256:1c15d6f31e2334c089da46b44939883879679608d006352547cb17a2b801d467"],"state_sha256":"eed12082df7e42c60c1c9814890fcf7c61a557259f384da9bd58dd1975ce1440"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"9GzAhOJtQK2vWIJYfj8eLpcLfTyBq7CizmJEuQOEIfzP4vGiMFNi1uBQu1StdmGoWR4WZ6TrTgUR9z1UZicoDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-30T17:27:05.919013Z","bundle_sha256":"114f266adeaee346028316b0b81cd49dda15c43470a9b0e632ad2959021d4551"}}