{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:XZIBOT557MKOZJU6F5YUB6ZIJX","short_pith_number":"pith:XZIBOT55","canonical_record":{"source":{"id":"2308.06394","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-11T21:35:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"30d3cbe548ca948264177c67e205fc127043e12d2efb3f65dae5a330f432b288","abstract_canon_sha256":"5eb01d13a10a7218153fa67929717676cf94eaa1f53df7107214220d78f79281"},"schema_version":"1.0"},"canonical_sha256":"be50174fbdfb14eca69e2f7140fb284df1bdbfe35aa2be1ab1dfcd1cc0db4383","source":{"kind":"arxiv","id":"2308.06394","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.06394","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"arxiv_version","alias_value":"2308.06394v3","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.06394","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_12","alias_value":"XZIBOT557MKO","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_16","alias_value":"XZIBOT557MKOZJU6","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_8","alias_value":"XZIBOT55","created_at":"2026-07-05T07:43:54Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:XZIBOT557MKOZJU6F5YUB6ZIJX","target":"record","payload":{"canonical_record":{"source":{"id":"2308.06394","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-11T21:35:20Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"30d3cbe548ca948264177c67e205fc127043e12d2efb3f65dae5a330f432b288","abstract_canon_sha256":"5eb01d13a10a7218153fa67929717676cf94eaa1f53df7107214220d78f79281"},"schema_version":"1.0"},"canonical_sha256":"be50174fbdfb14eca69e2f7140fb284df1bdbfe35aa2be1ab1dfcd1cc0db4383","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:43:54.880432Z","signature_b64":"qlexFed2TGctVTXaucS1umkq7YphZIJfNIUxbEI2r6dMQS0DXjoMbxfGCpAQ4Nyh2ziPCpFBVIeYOoO//vbGCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"be50174fbdfb14eca69e2f7140fb284df1bdbfe35aa2be1ab1dfcd1cc0db4383","last_reissued_at":"2026-07-05T07:43:54.879995Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:43:54.879995Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2308.06394","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T07:43:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3WUqMwEeqezu1vfhw/l22TZT0nFz4Br4NH6nSeUKVRFhFtw1i/R4q7QZvHq4OiSEpFb444njW+FnW6+aftvLDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T02:25:06.329187Z"},"content_sha256":"c016a9a6a80e709a9881deae0c9e7ce66dd7aa1782e527227d8f4761e6a0b432","schema_version":"1.0","event_id":"sha256:c016a9a6a80e709a9881deae0c9e7ce66dd7aa1782e527227d8f4761e6a0b432"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:XZIBOT557MKOZJU6F5YUB6ZIJX","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Detecting and Preventing Hallucinations in Large Vision Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Anisha Gunjal, Erhan Bas, Jihan Yin","submitted_at":"2023-08-11T21:35:20Z","abstract_excerpt":"Instruction tuned Large Vision Language Models (LVLMs) have significantly advanced in generalizing across a diverse set of multi-modal tasks, especially for Visual Question Answering (VQA). However, generating detailed responses that are visually grounded is still a challenging task for these models. We find that even the current state-of-the-art LVLMs (InstructBLIP) still contain a staggering 30 percent of the hallucinatory text in the form of non-existent objects, unfaithful descriptions, and inaccurate relationships. To address this, we introduce M-HalDetect, a (M)ultimodal (Hal)lucination "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.06394","kind":"arxiv","version":3},"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/2308.06394/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":"2026-07-05T07:43:54Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H0HiHB+WFE4RB7qvbuQ0YBvrQZg69QYzJnqQCjURvX+Fr3DZvUh86GQC5JSOCA75crFVY1lzpAnr4RlZ74BTAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-05T02:25:06.329678Z"},"content_sha256":"dd767acdfaa496cea7b529273156c71755def17a7feff1a646fdab0c4294630b","schema_version":"1.0","event_id":"sha256:dd767acdfaa496cea7b529273156c71755def17a7feff1a646fdab0c4294630b"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/bundle.json","state_url":"https://pith.science/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/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":"2026-08-05T02:25:06Z","links":{"resolver":"https://pith.science/pith/XZIBOT557MKOZJU6F5YUB6ZIJX","bundle":"https://pith.science/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/bundle.json","state":"https://pith.science/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/state.json","well_known_bundle":"https://pith.science/.well-known/pith/XZIBOT557MKOZJU6F5YUB6ZIJX/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:XZIBOT557MKOZJU6F5YUB6ZIJX","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":"5eb01d13a10a7218153fa67929717676cf94eaa1f53df7107214220d78f79281","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-11T21:35:20Z","title_canon_sha256":"30d3cbe548ca948264177c67e205fc127043e12d2efb3f65dae5a330f432b288"},"schema_version":"1.0","source":{"id":"2308.06394","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2308.06394","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"arxiv_version","alias_value":"2308.06394v3","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2308.06394","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_12","alias_value":"XZIBOT557MKO","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_16","alias_value":"XZIBOT557MKOZJU6","created_at":"2026-07-05T07:43:54Z"},{"alias_kind":"pith_short_8","alias_value":"XZIBOT55","created_at":"2026-07-05T07:43:54Z"}],"graph_snapshots":[{"event_id":"sha256:dd767acdfaa496cea7b529273156c71755def17a7feff1a646fdab0c4294630b","target":"graph","created_at":"2026-07-05T07:43:54Z","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/2308.06394/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction tuned Large Vision Language Models (LVLMs) have significantly advanced in generalizing across a diverse set of multi-modal tasks, especially for Visual Question Answering (VQA). However, generating detailed responses that are visually grounded is still a challenging task for these models. We find that even the current state-of-the-art LVLMs (InstructBLIP) still contain a staggering 30 percent of the hallucinatory text in the form of non-existent objects, unfaithful descriptions, and inaccurate relationships. To address this, we introduce M-HalDetect, a (M)ultimodal (Hal)lucination ","authors_text":"Anisha Gunjal, Erhan Bas, Jihan Yin","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-11T21:35:20Z","title":"Detecting and Preventing Hallucinations in Large Vision Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2308.06394","kind":"arxiv","version":3},"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:c016a9a6a80e709a9881deae0c9e7ce66dd7aa1782e527227d8f4761e6a0b432","target":"record","created_at":"2026-07-05T07:43:54Z","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":"5eb01d13a10a7218153fa67929717676cf94eaa1f53df7107214220d78f79281","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2023-08-11T21:35:20Z","title_canon_sha256":"30d3cbe548ca948264177c67e205fc127043e12d2efb3f65dae5a330f432b288"},"schema_version":"1.0","source":{"id":"2308.06394","kind":"arxiv","version":3}},"canonical_sha256":"be50174fbdfb14eca69e2f7140fb284df1bdbfe35aa2be1ab1dfcd1cc0db4383","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"be50174fbdfb14eca69e2f7140fb284df1bdbfe35aa2be1ab1dfcd1cc0db4383","first_computed_at":"2026-07-05T07:43:54.879995Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:43:54.879995Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"qlexFed2TGctVTXaucS1umkq7YphZIJfNIUxbEI2r6dMQS0DXjoMbxfGCpAQ4Nyh2ziPCpFBVIeYOoO//vbGCQ==","signature_status":"signed_v1","signed_at":"2026-07-05T07:43:54.880432Z","signed_message":"canonical_sha256_bytes"},"source_id":"2308.06394","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c016a9a6a80e709a9881deae0c9e7ce66dd7aa1782e527227d8f4761e6a0b432","sha256:dd767acdfaa496cea7b529273156c71755def17a7feff1a646fdab0c4294630b"],"state_sha256":"3cdb091c58fea69d4c888a3014df50f6ec8ab1e3b5229eb63fe220c8421c8240"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ILpUFr+KNHH99vzJnk0EiyN6tpbRfRFknUSe8jS43rXppfHbUEg8q3wQu5fIUyOw0BTncizkOPy24uJ1fMUNBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-05T02:25:06.334088Z","bundle_sha256":"adcca355c13df1237583333cb4f20af882d2448337173255e34beffcc398693e"}}