{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ","short_pith_number":"pith:T3YRJ3P2","canonical_record":{"source":{"id":"2403.12145","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T18:01:26Z","cross_cats_sorted":[],"title_canon_sha256":"bd1441cec12d645b9c91508ce0947db57c6c7cbe25e3873ed9e1e192fffa3cdd","abstract_canon_sha256":"c7a68922787b0ed70cfb3acc11b7b531bc77515937b7989fdece2e7be218a21e"},"schema_version":"1.0"},"canonical_sha256":"9ef114edfac9478cf308c39541d2f2ce69f0fe8a019fe2e217de60564774c3b3","source":{"kind":"arxiv","id":"2403.12145","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12145","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12145v1","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12145","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_12","alias_value":"T3YRJ3P2ZFDY","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_16","alias_value":"T3YRJ3P2ZFDYZ4YI","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_8","alias_value":"T3YRJ3P2","created_at":"2026-07-05T07:57:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ","target":"record","payload":{"canonical_record":{"source":{"id":"2403.12145","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T18:01:26Z","cross_cats_sorted":[],"title_canon_sha256":"bd1441cec12d645b9c91508ce0947db57c6c7cbe25e3873ed9e1e192fffa3cdd","abstract_canon_sha256":"c7a68922787b0ed70cfb3acc11b7b531bc77515937b7989fdece2e7be218a21e"},"schema_version":"1.0"},"canonical_sha256":"9ef114edfac9478cf308c39541d2f2ce69f0fe8a019fe2e217de60564774c3b3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:57:55.976116Z","signature_b64":"bufJ104wS//D4XsD2grx2dhXusyvJUiAgekI66Q69SD4Ics+suQl5HxuN/eRWFFZrKXj+s8RHQ4T1QKjEkRJCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9ef114edfac9478cf308c39541d2f2ce69f0fe8a019fe2e217de60564774c3b3","last_reissued_at":"2026-07-05T07:57:55.975525Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:57:55.975525Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2403.12145","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":"2026-07-05T07:57:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"yaWnvmiSiz+ULxI9wkSXPGIETW2p2tlY2YseZ8jl0i3zyCgM3uchKR0OhECBX6UMKgpVYSD/M+swSRQA20cnAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:38:37.559736Z"},"content_sha256":"5936c33d5aaeb0e6ad9fe3f8584080fff2e157c212b4e9b046121ee0d545e1d2","schema_version":"1.0","event_id":"sha256:5936c33d5aaeb0e6ad9fe3f8584080fff2e157c212b4e9b046121ee0d545e1d2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Syn-QA2: Evaluating False Assumptions in Long-tail Questions with Synthetic QA Datasets","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ashwin Daswani, Najoung Kim, Rohan Sawant","submitted_at":"2024-03-18T18:01:26Z","abstract_excerpt":"Sensitivity to false assumptions (or false premises) in information-seeking questions is critical for robust question-answering (QA) systems. Recent work has shown that false assumptions in naturally occurring questions pose challenges to current models, with low performance on both generative QA and simple detection tasks (Kim et al. 2023). However, the focus of existing work on naturally occurring questions leads to a gap in the analysis of model behavior on the long tail of the distribution of possible questions. To this end, we introduce Syn-(QA)$^2$, a set of two synthetically generated Q"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12145","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/2403.12145/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":"2026-07-05T07:57:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2Hvlk9h6qrpHhnfUySdGwKG/oWPSPokXGqvB4j3+CRbbqPltJUoRQnedIAb/CZIkad5CKsBWl5W7TRfrrOWgBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T03:38:37.560367Z"},"content_sha256":"14e550e2141de7a004d0510508fc50d0b37b177902aaaf80878b0b709f1bca4a","schema_version":"1.0","event_id":"sha256:14e550e2141de7a004d0510508fc50d0b37b177902aaaf80878b0b709f1bca4a"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/bundle.json","state_url":"https://pith.science/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/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":"2026-08-08T03:38:37Z","links":{"resolver":"https://pith.science/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ","bundle":"https://pith.science/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/bundle.json","state":"https://pith.science/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/state.json","well_known_bundle":"https://pith.science/.well-known/pith/T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:T3YRJ3P2ZFDYZ4YIYOKUDUXSZZ","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":"c7a68922787b0ed70cfb3acc11b7b531bc77515937b7989fdece2e7be218a21e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T18:01:26Z","title_canon_sha256":"bd1441cec12d645b9c91508ce0947db57c6c7cbe25e3873ed9e1e192fffa3cdd"},"schema_version":"1.0","source":{"id":"2403.12145","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2403.12145","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"arxiv_version","alias_value":"2403.12145v1","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2403.12145","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_12","alias_value":"T3YRJ3P2ZFDY","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_16","alias_value":"T3YRJ3P2ZFDYZ4YI","created_at":"2026-07-05T07:57:55Z"},{"alias_kind":"pith_short_8","alias_value":"T3YRJ3P2","created_at":"2026-07-05T07:57:55Z"}],"graph_snapshots":[{"event_id":"sha256:14e550e2141de7a004d0510508fc50d0b37b177902aaaf80878b0b709f1bca4a","target":"graph","created_at":"2026-07-05T07:57:55Z","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/2403.12145/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Sensitivity to false assumptions (or false premises) in information-seeking questions is critical for robust question-answering (QA) systems. Recent work has shown that false assumptions in naturally occurring questions pose challenges to current models, with low performance on both generative QA and simple detection tasks (Kim et al. 2023). However, the focus of existing work on naturally occurring questions leads to a gap in the analysis of model behavior on the long tail of the distribution of possible questions. To this end, we introduce Syn-(QA)$^2$, a set of two synthetically generated Q","authors_text":"Ashwin Daswani, Najoung Kim, Rohan Sawant","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T18:01:26Z","title":"Syn-QA2: Evaluating False Assumptions in Long-tail Questions with Synthetic QA Datasets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2403.12145","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":"sha256:5936c33d5aaeb0e6ad9fe3f8584080fff2e157c212b4e9b046121ee0d545e1d2","target":"record","created_at":"2026-07-05T07:57:55Z","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":"c7a68922787b0ed70cfb3acc11b7b531bc77515937b7989fdece2e7be218a21e","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-03-18T18:01:26Z","title_canon_sha256":"bd1441cec12d645b9c91508ce0947db57c6c7cbe25e3873ed9e1e192fffa3cdd"},"schema_version":"1.0","source":{"id":"2403.12145","kind":"arxiv","version":1}},"canonical_sha256":"9ef114edfac9478cf308c39541d2f2ce69f0fe8a019fe2e217de60564774c3b3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9ef114edfac9478cf308c39541d2f2ce69f0fe8a019fe2e217de60564774c3b3","first_computed_at":"2026-07-05T07:57:55.975525Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:57:55.975525Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bufJ104wS//D4XsD2grx2dhXusyvJUiAgekI66Q69SD4Ics+suQl5HxuN/eRWFFZrKXj+s8RHQ4T1QKjEkRJCg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:57:55.976116Z","signed_message":"canonical_sha256_bytes"},"source_id":"2403.12145","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:5936c33d5aaeb0e6ad9fe3f8584080fff2e157c212b4e9b046121ee0d545e1d2","sha256:14e550e2141de7a004d0510508fc50d0b37b177902aaaf80878b0b709f1bca4a"],"state_sha256":"a4330fe70e4df32e2972a713509ed7a871478e9ee62e372d108a88734a1d8f7f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xeUumNciT3Gi7WurYY50k2lpNAF9ztZWZ69nJ/o7qsGV2Ksy+XbENF6zERKgijeO3pCd05oJ0ItemHvgZqbrBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T03:38:37.564587Z","bundle_sha256":"de379ee10a31a4ecf4dcf14ec3958e49e9458d5653561e877ecdb0b5101c473e"}}