{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:U3FOD42Y2AP7VYVIEYBQDP4FTU","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":"4a01de5801cd0fc5dd2b65921800c1c0dad49b766486ac0de397a27d1bfa8189","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-03T23:03:24Z","title_canon_sha256":"0de08e1270f53cee3b1fe0897907fa745c17e263b9e193f0d5c36e01106d879d"},"schema_version":"1.0","source":{"id":"2501.02135","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2501.02135","created_at":"2026-07-05T09:57:02Z"},{"alias_kind":"arxiv_version","alias_value":"2501.02135v1","created_at":"2026-07-05T09:57:02Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.02135","created_at":"2026-07-05T09:57:02Z"},{"alias_kind":"pith_short_12","alias_value":"U3FOD42Y2AP7","created_at":"2026-07-05T09:57:02Z"},{"alias_kind":"pith_short_16","alias_value":"U3FOD42Y2AP7VYVI","created_at":"2026-07-05T09:57:02Z"},{"alias_kind":"pith_short_8","alias_value":"U3FOD42Y","created_at":"2026-07-05T09:57:02Z"}],"graph_snapshots":[{"event_id":"sha256:56621bd3f5640f03b6cf11ba0a1939f3777d137558c0b7dd59764f4dcff74b88","target":"graph","created_at":"2026-07-05T09:57:02Z","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/2501.02135/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the rapid advancement of Multi-modal Large Language Models (MLLMs), several diagnostic benchmarks have recently been developed to assess these models' multi-modal reasoning proficiency. However, these benchmarks are restricted to assessing primarily the visual aspect and do not examine the holistic audio-visual (AV) understanding. Moreover, currently, there are no benchmarks that investigate the capabilities of AVLLMs to calibrate their responses when presented with perturbed inputs. To this end, we introduce Audio-Visual Trustworthiness assessment Benchmark (AVTrustBench), comprising 600","authors_text":"Dinesh Manocha, Mohamed Elhoseiny, Ruohan Gao, Sanjoy Chowdhury, Sayan Nag, Subhrajyoti Dasgupta, Yaoting Wang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-03T23:03:24Z","title":"AVTrustBench: Assessing and Enhancing Reliability and Robustness in Audio-Visual LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.02135","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":"sha256:8e39c07446a62cad99f23eb5a167beff2bdb279f6c3c3559aa2cd632fa200c88","target":"record","created_at":"2026-07-05T09:57:02Z","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":"4a01de5801cd0fc5dd2b65921800c1c0dad49b766486ac0de397a27d1bfa8189","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-03T23:03:24Z","title_canon_sha256":"0de08e1270f53cee3b1fe0897907fa745c17e263b9e193f0d5c36e01106d879d"},"schema_version":"1.0","source":{"id":"2501.02135","kind":"arxiv","version":1}},"canonical_sha256":"a6cae1f358d01ffae2a8260301bf859d2595c09afe201f86142439c184d8d056","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"a6cae1f358d01ffae2a8260301bf859d2595c09afe201f86142439c184d8d056","first_computed_at":"2026-07-05T09:57:02.700809Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:57:02.700809Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ljwx84C+cJ9RrgAfHfmIfnmMor5OhKuUnY3ZPk7S1MEGlmcDtPXt4YsC/gp+myZNvJDekFz/goRo3+IYIOiEBg==","signature_status":"signed_v1","signed_at":"2026-07-05T09:57:02.701226Z","signed_message":"canonical_sha256_bytes"},"source_id":"2501.02135","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8e39c07446a62cad99f23eb5a167beff2bdb279f6c3c3559aa2cd632fa200c88","sha256:56621bd3f5640f03b6cf11ba0a1939f3777d137558c0b7dd59764f4dcff74b88"],"state_sha256":"b194504b7fb7ae2de5cd6a6e768043331d411bdaba8aab018be9e39951a15520"}