{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:NWP3R6ZOOYP4P6QAXQ4644XGPQ","short_pith_number":"pith:NWP3R6ZO","canonical_record":{"source":{"id":"2403.20331","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-29T17:59:53Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"37c1b8127c648fc85633c8f4946397e4c0feaa545bbe4df0d27667c38aa73f04","abstract_canon_sha256":"3c50f2e1e536479ee647d0193464ff97961b27c4753a32865a427431bc4c3e6e"},"schema_version":"1.0"},"canonical_sha256":"6d9fb8fb2e761fc7fa00bc39ee72e67c27cfd74af55f9a0b423e5d505c91f67d","source":{"kind":"arxiv","id":"2403.20331","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.20331","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"arxiv_version","alias_value":"2403.20331v4","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.20331","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_12","alias_value":"NWP3R6ZOOYP4","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_16","alias_value":"NWP3R6ZOOYP4P6QA","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_8","alias_value":"NWP3R6ZO","created_at":"2026-07-05T11:18:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:NWP3R6ZOOYP4P6QAXQ4644XGPQ","target":"record","payload":{"canonical_record":{"source":{"id":"2403.20331","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-29T17:59:53Z","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"title_canon_sha256":"37c1b8127c648fc85633c8f4946397e4c0feaa545bbe4df0d27667c38aa73f04","abstract_canon_sha256":"3c50f2e1e536479ee647d0193464ff97961b27c4753a32865a427431bc4c3e6e"},"schema_version":"1.0"},"canonical_sha256":"6d9fb8fb2e761fc7fa00bc39ee72e67c27cfd74af55f9a0b423e5d505c91f67d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:18:11.750694Z","signature_b64":"Nz0aWP5qK3uF/ItdNTTxtU4B27iLyO/xwtF9t6LtYAZnraNnEJHjrGixtc/lZlsIIPC5PwD7g2Hms9rTMHDmCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6d9fb8fb2e761fc7fa00bc39ee72e67c27cfd74af55f9a0b423e5d505c91f67d","last_reissued_at":"2026-07-05T11:18:11.750149Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:18:11.750149Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.20331","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:18:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"LIlpgHWijIFuJeybASEvGD/aSzn+LKt2XiHQ8z7ijouWYQMtn8zlbnBSKYiS+/0F5UcB85yN+40i99ZFf9cBBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:03:12.009949Z"},"content_sha256":"161dea25a4810a6ef391d2b4e638a9aa0eb935e85c65a1cffc9073a74450e862","schema_version":"1.0","event_id":"sha256:161dea25a4810a6ef391d2b4e638a9aa0eb935e85c65a1cffc9073a74450e862"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:NWP3R6ZOOYP4P6QAXQ4644XGPQ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Unsolvable Problem Detection: Robust Understanding Evaluation for Large Multimodal Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CL","cs.LG"],"primary_cat":"cs.CV","authors_text":"Atsuyuki Miyai, Go Irie, Hai Li, Jingkang Yang, Jingyang Zhang, Kiyoharu Aizawa, Qing Yu, Yifei Ming, Yixuan Li, Ziwei Liu","submitted_at":"2024-03-29T17:59:53Z","abstract_excerpt":"This paper introduces a novel task to evaluate the robust understanding capability of Large Multimodal Models (LMMs), termed $\\textbf{Unsolvable Problem Detection (UPD)}$. Multiple-choice question answering (MCQA) is widely used to assess the understanding capability of LMMs, but it does not guarantee that LMMs truly comprehend the answer. UPD assesses the LMM's ability to withhold answers when encountering unsolvable problems of MCQA, verifying whether the model truly understands the answer. UPD encompasses three problems: Absent Answer Detection (AAD), Incompatible Answer Set Detection (IASD"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.20331","kind":"arxiv","version":4},"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/2403.20331/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":"2026-07-05T11:18:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"38HkdzcRJBe2+7X61ykqufIZ4xh7fkYuEYynXU1ZmGsT4HsAIHqgvR8rY4kgAUSgH/NdEO4likpIjTUZVuqBDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T08:03:12.010466Z"},"content_sha256":"a66fda9d6d8f88b1eaf1ee1dc146c37211839d013bedbb9c3bbc110174a9b106","schema_version":"1.0","event_id":"sha256:a66fda9d6d8f88b1eaf1ee1dc146c37211839d013bedbb9c3bbc110174a9b106"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/bundle.json","state_url":"https://pith.science/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/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":"2026-08-08T08:03:12Z","links":{"resolver":"https://pith.science/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ","bundle":"https://pith.science/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/bundle.json","state":"https://pith.science/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NWP3R6ZOOYP4P6QAXQ4644XGPQ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:NWP3R6ZOOYP4P6QAXQ4644XGPQ","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":"3c50f2e1e536479ee647d0193464ff97961b27c4753a32865a427431bc4c3e6e","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-29T17:59:53Z","title_canon_sha256":"37c1b8127c648fc85633c8f4946397e4c0feaa545bbe4df0d27667c38aa73f04"},"schema_version":"1.0","source":{"id":"2403.20331","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.20331","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"arxiv_version","alias_value":"2403.20331v4","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.20331","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_12","alias_value":"NWP3R6ZOOYP4","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_16","alias_value":"NWP3R6ZOOYP4P6QA","created_at":"2026-07-05T11:18:11Z"},{"alias_kind":"pith_short_8","alias_value":"NWP3R6ZO","created_at":"2026-07-05T11:18:11Z"}],"graph_snapshots":[{"event_id":"sha256:a66fda9d6d8f88b1eaf1ee1dc146c37211839d013bedbb9c3bbc110174a9b106","target":"graph","created_at":"2026-07-05T11:18:11Z","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/2403.20331/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces a novel task to evaluate the robust understanding capability of Large Multimodal Models (LMMs), termed $\\textbf{Unsolvable Problem Detection (UPD)}$. Multiple-choice question answering (MCQA) is widely used to assess the understanding capability of LMMs, but it does not guarantee that LMMs truly comprehend the answer. UPD assesses the LMM's ability to withhold answers when encountering unsolvable problems of MCQA, verifying whether the model truly understands the answer. UPD encompasses three problems: Absent Answer Detection (AAD), Incompatible Answer Set Detection (IASD","authors_text":"Atsuyuki Miyai, Go Irie, Hai Li, Jingkang Yang, Jingyang Zhang, Kiyoharu Aizawa, Qing Yu, Yifei Ming, Yixuan Li, Ziwei Liu","cross_cats":["cs.AI","cs.CL","cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-29T17:59:53Z","title":"Unsolvable Problem Detection: Robust Understanding Evaluation for Large Multimodal Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.20331","kind":"arxiv","version":4},"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":"sha256:161dea25a4810a6ef391d2b4e638a9aa0eb935e85c65a1cffc9073a74450e862","target":"record","created_at":"2026-07-05T11:18:11Z","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":"3c50f2e1e536479ee647d0193464ff97961b27c4753a32865a427431bc4c3e6e","cross_cats_sorted":["cs.AI","cs.CL","cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-29T17:59:53Z","title_canon_sha256":"37c1b8127c648fc85633c8f4946397e4c0feaa545bbe4df0d27667c38aa73f04"},"schema_version":"1.0","source":{"id":"2403.20331","kind":"arxiv","version":4}},"canonical_sha256":"6d9fb8fb2e761fc7fa00bc39ee72e67c27cfd74af55f9a0b423e5d505c91f67d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6d9fb8fb2e761fc7fa00bc39ee72e67c27cfd74af55f9a0b423e5d505c91f67d","first_computed_at":"2026-07-05T11:18:11.750149Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:18:11.750149Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Nz0aWP5qK3uF/ItdNTTxtU4B27iLyO/xwtF9t6LtYAZnraNnEJHjrGixtc/lZlsIIPC5PwD7g2Hms9rTMHDmCw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:18:11.750694Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.20331","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:161dea25a4810a6ef391d2b4e638a9aa0eb935e85c65a1cffc9073a74450e862","sha256:a66fda9d6d8f88b1eaf1ee1dc146c37211839d013bedbb9c3bbc110174a9b106"],"state_sha256":"83934555042fd010b64ad6d961047904f521076c843a5350367e7104026f1c2c"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KKNBhvKn3c7nqrzW2dWeem62BH9F1MGu8TL5U8yEZ3uqmbwJCwxCBgJbQoK5ZWXVLDf+nTdId2NQTQVrUSi/Cg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T08:03:12.015753Z","bundle_sha256":"550f97dbec3a04673a56a2566d027641fcd742f0f421ba8a3c81ff43f63272bb"}}