{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:JS4M2AYZFPYABAXZZ3PRQG2ZP7","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":"6118f9d36a2f38f7e4a8bb8c5466d318ef0dda186be38020f447e17d7d5fe4ee","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-03T06:25:45Z","title_canon_sha256":"df4ffbc0418d9e796aa3ec3316389184bfe68c5ec5024de7d1ffb96cc6f2e5d6"},"schema_version":"1.0","source":{"id":"2605.08145","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.08145","created_at":"2026-06-01T01:02:42Z"},{"alias_kind":"arxiv_version","alias_value":"2605.08145v2","created_at":"2026-06-01T01:02:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.08145","created_at":"2026-06-01T01:02:42Z"},{"alias_kind":"pith_short_12","alias_value":"JS4M2AYZFPYA","created_at":"2026-06-01T01:02:42Z"},{"alias_kind":"pith_short_16","alias_value":"JS4M2AYZFPYABAXZ","created_at":"2026-06-01T01:02:42Z"},{"alias_kind":"pith_short_8","alias_value":"JS4M2AYZ","created_at":"2026-06-01T01:02:42Z"}],"graph_snapshots":[{"event_id":"sha256:8a6b51e26a5347bddd36276576253c80f69f5d23f990969774647a1144b057eb","target":"graph","created_at":"2026-06-01T01:02:42Z","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":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Our findings suggest that increasing redundancy can reduce visual induced errors by 38.3% and improve consistency by 16.8%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The assumption that modern instruction datasets eliminate redundancies to prioritize visual grounding and that converting unique interactions to redundant ones via the Multimodal Interaction Gate will reliably compensate for impaired modalities without introducing new failure modes or losing synergistic information."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A self-captioning method using a Multimodal Interaction Gate amplifies redundant interactions to reduce visual-induced errors by 38.3% and improve consistency by 16.8% in vision-language models."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Amplifying redundant multimodal interactions reduces visual errors in vision-language models by 38.3%."}],"snapshot_sha256":"a6d0b2b7a5aae4a7a8c1d047e2fe53576b716862edb2ddbdd5642cdcf3b1926d"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"76198005f0708c21cc5864a71ca7606e2c5b7ed2a15face8447a69dbc2557e4d"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T17:37:16.901423Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-20T05:01:23.062289Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T17:00:49.982953Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.08145/integrity.json","findings":[],"snapshot_sha256":"ca6149d1a8ff6caa6da4a950be6d7b3515ea1314bbfccda9d64b19fc6301d21f","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Current vision language models face hallucination and robustness issues against ambiguous or corrupted modalities. We hypothesize that these issues can be addressed by exploiting the shared information between modalities to compensate for the impaired one. To this end, we analyze multimodal interactions -- redundant (shared), unique (exclusive), and synergistic (emergent) task-relevant information provided by the modalities -- to determine their impacts on model reliability. Specifically, amplifying redundant interactions would increase this exploitable shared information to resolve these issu","authors_text":"Adriel Kuek, Hei Man Ip, Paul Pu Liang, Roy Ka-Wei Lee, Yuriel Ryan","cross_cats":["cs.AI","cs.LG"],"headline":"Amplifying redundant multimodal interactions reduces visual errors in vision-language models by 38.3%.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-03T06:25:45Z","title":"Self-Captioning Multimodal Interaction Tuning: Amplifying Exploitable Redundancies for Robust Vision Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.08145","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T01:33:43.764074Z","id":"78cb1e18-87c3-401f-a725-c00afb9916fb","model_set":{"reader":"grok-4.3"},"one_line_summary":"A self-captioning method using a Multimodal Interaction Gate amplifies redundant interactions to reduce visual-induced errors by 38.3% and improve consistency by 16.8% in vision-language models.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Amplifying redundant multimodal interactions reduces visual errors in vision-language models by 38.3%.","strongest_claim":"Our findings suggest that increasing redundancy can reduce visual induced errors by 38.3% and improve consistency by 16.8%.","weakest_assumption":"The assumption that modern instruction datasets eliminate redundancies to prioritize visual grounding and that converting unique interactions to redundant ones via the Multimodal Interaction Gate will reliably compensate for impaired modalities without introducing new failure modes or losing synergistic information."}},"verdict_id":"78cb1e18-87c3-401f-a725-c00afb9916fb"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:6cdb1219e870c558ec78745e6ff97b7a1e3f8c8157824f44866f0e1cc63e41d9","target":"record","created_at":"2026-06-01T01:02:42Z","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":"6118f9d36a2f38f7e4a8bb8c5466d318ef0dda186be38020f447e17d7d5fe4ee","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-05-03T06:25:45Z","title_canon_sha256":"df4ffbc0418d9e796aa3ec3316389184bfe68c5ec5024de7d1ffb96cc6f2e5d6"},"schema_version":"1.0","source":{"id":"2605.08145","kind":"arxiv","version":2}},"canonical_sha256":"4cb8cd03192bf00082f9cedf181b597fcaa6223af31df2c0c0c4c3810ccdfeba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4cb8cd03192bf00082f9cedf181b597fcaa6223af31df2c0c0c4c3810ccdfeba","first_computed_at":"2026-06-01T01:02:42.331369Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-01T01:02:42.331369Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JN5wQ4NjfyQkSKxfs9ykbsVa90Dus0Y5+aXTHcVZq8In46SFAHzdks7MwWsSvH1MpAqHBai1MMdmz6UE3soWAg==","signature_status":"signed_v1","signed_at":"2026-06-01T01:02:42.332288Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.08145","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:6cdb1219e870c558ec78745e6ff97b7a1e3f8c8157824f44866f0e1cc63e41d9","sha256:8a6b51e26a5347bddd36276576253c80f69f5d23f990969774647a1144b057eb"],"state_sha256":"d580815007e6757f883eaa1934d7812f7f649cee0dc18cc9599720bfe843c0b2"}