{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:FRVSB3SLE7QE4JOSZSAX73U424","short_pith_number":"pith:FRVSB3SL","canonical_record":{"source":{"id":"2404.12452","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T18:26:43Z","cross_cats_sorted":[],"title_canon_sha256":"8735366f025470933b1d6dfc22007aaa0cdaf86faaba6e7ab22a94a5b57e7bd7","abstract_canon_sha256":"8fb2ff8e069610af03093c1ac8d843fa87406d7d91522b1b6d491371f40f1490"},"schema_version":"1.0"},"canonical_sha256":"2c6b20ee4b27e04e25d2cc817fee9cd714a5c686629b4a0c85c69b05afb0824d","source":{"kind":"arxiv","id":"2404.12452","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.12452","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"arxiv_version","alias_value":"2404.12452v2","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.12452","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_12","alias_value":"FRVSB3SLE7QE","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_16","alias_value":"FRVSB3SLE7QE4JOS","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_8","alias_value":"FRVSB3SL","created_at":"2026-07-05T09:16:14Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:FRVSB3SLE7QE4JOSZSAX73U424","target":"record","payload":{"canonical_record":{"source":{"id":"2404.12452","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T18:26:43Z","cross_cats_sorted":[],"title_canon_sha256":"8735366f025470933b1d6dfc22007aaa0cdaf86faaba6e7ab22a94a5b57e7bd7","abstract_canon_sha256":"8fb2ff8e069610af03093c1ac8d843fa87406d7d91522b1b6d491371f40f1490"},"schema_version":"1.0"},"canonical_sha256":"2c6b20ee4b27e04e25d2cc817fee9cd714a5c686629b4a0c85c69b05afb0824d","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:16:14.525336Z","signature_b64":"2snOmQ3X39cqws8kCw2Fkd/t9zLy0vW4Ub6OLW16hk+lJemgwD2rVw7AR4dcH0pS/00KX8FSQdLmQXg9OHTjAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2c6b20ee4b27e04e25d2cc817fee9cd714a5c686629b4a0c85c69b05afb0824d","last_reissued_at":"2026-07-05T09:16:14.524926Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:16:14.524926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.12452","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T09:16:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FnVSPsFpUjWMFTMfwjcHtXP6WxDDJDy8WmuXsTlerv++UBqsDCXY6eBs/RREJ+HY+ZJhX+6Ihe/brxIw4X7AAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:03:00.389532Z"},"content_sha256":"c36bc98d110e7bf5c5b5ea780765d705019fe82964f91d619d999e813bf2a5d3","schema_version":"1.0","event_id":"sha256:c36bc98d110e7bf5c5b5ea780765d705019fe82964f91d619d999e813bf2a5d3"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:FRVSB3SLE7QE4JOSZSAX73U424","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Characterizing LLM Abstention Behavior in Science QA with Context Perturbations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Bill Howe, Bingbing Wen, Lucy Lu Wang","submitted_at":"2024-04-18T18:26:43Z","abstract_excerpt":"The correct model response in the face of uncertainty is to abstain from answering a question so as not to mislead the user. In this work, we study the ability of LLMs to abstain from answering context-dependent science questions when provided insufficient or incorrect context. We probe model sensitivity in several settings: removing gold context, replacing gold context with irrelevant context, and providing additional context beyond what is given. In experiments on four QA datasets with six LLMs, we show that performance varies greatly across models, across the type of context provided, and a"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.12452","kind":"arxiv","version":2},"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/2404.12452/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":"2026-07-05T09:16:14Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"I+xQmpK5SP2aQ6VL5608e4VsUg7mk4smUpXwL/lN9uGoyQichBRJJMoNLfslnMfD/bM2DrTVoYd+PIQuyvYTAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:03:00.390427Z"},"content_sha256":"78670bd90de9cac47f6b9cfbc5b39b5657202d102f74a2bba635a7c2712a045c","schema_version":"1.0","event_id":"sha256:78670bd90de9cac47f6b9cfbc5b39b5657202d102f74a2bba635a7c2712a045c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FRVSB3SLE7QE4JOSZSAX73U424/bundle.json","state_url":"https://pith.science/pith/FRVSB3SLE7QE4JOSZSAX73U424/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FRVSB3SLE7QE4JOSZSAX73U424/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":"2026-08-06T09:03:00Z","links":{"resolver":"https://pith.science/pith/FRVSB3SLE7QE4JOSZSAX73U424","bundle":"https://pith.science/pith/FRVSB3SLE7QE4JOSZSAX73U424/bundle.json","state":"https://pith.science/pith/FRVSB3SLE7QE4JOSZSAX73U424/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FRVSB3SLE7QE4JOSZSAX73U424/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:FRVSB3SLE7QE4JOSZSAX73U424","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":"8fb2ff8e069610af03093c1ac8d843fa87406d7d91522b1b6d491371f40f1490","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T18:26:43Z","title_canon_sha256":"8735366f025470933b1d6dfc22007aaa0cdaf86faaba6e7ab22a94a5b57e7bd7"},"schema_version":"1.0","source":{"id":"2404.12452","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.12452","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"arxiv_version","alias_value":"2404.12452v2","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.12452","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_12","alias_value":"FRVSB3SLE7QE","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_16","alias_value":"FRVSB3SLE7QE4JOS","created_at":"2026-07-05T09:16:14Z"},{"alias_kind":"pith_short_8","alias_value":"FRVSB3SL","created_at":"2026-07-05T09:16:14Z"}],"graph_snapshots":[{"event_id":"sha256:78670bd90de9cac47f6b9cfbc5b39b5657202d102f74a2bba635a7c2712a045c","target":"graph","created_at":"2026-07-05T09:16:14Z","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/2404.12452/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The correct model response in the face of uncertainty is to abstain from answering a question so as not to mislead the user. In this work, we study the ability of LLMs to abstain from answering context-dependent science questions when provided insufficient or incorrect context. We probe model sensitivity in several settings: removing gold context, replacing gold context with irrelevant context, and providing additional context beyond what is given. In experiments on four QA datasets with six LLMs, we show that performance varies greatly across models, across the type of context provided, and a","authors_text":"Bill Howe, Bingbing Wen, Lucy Lu Wang","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T18:26:43Z","title":"Characterizing LLM Abstention Behavior in Science QA with Context Perturbations"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.12452","kind":"arxiv","version":2},"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:c36bc98d110e7bf5c5b5ea780765d705019fe82964f91d619d999e813bf2a5d3","target":"record","created_at":"2026-07-05T09:16:14Z","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":"8fb2ff8e069610af03093c1ac8d843fa87406d7d91522b1b6d491371f40f1490","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-18T18:26:43Z","title_canon_sha256":"8735366f025470933b1d6dfc22007aaa0cdaf86faaba6e7ab22a94a5b57e7bd7"},"schema_version":"1.0","source":{"id":"2404.12452","kind":"arxiv","version":2}},"canonical_sha256":"2c6b20ee4b27e04e25d2cc817fee9cd714a5c686629b4a0c85c69b05afb0824d","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2c6b20ee4b27e04e25d2cc817fee9cd714a5c686629b4a0c85c69b05afb0824d","first_computed_at":"2026-07-05T09:16:14.524926Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:16:14.524926Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2snOmQ3X39cqws8kCw2Fkd/t9zLy0vW4Ub6OLW16hk+lJemgwD2rVw7AR4dcH0pS/00KX8FSQdLmQXg9OHTjAA==","signature_status":"signed_v1","signed_at":"2026-07-05T09:16:14.525336Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.12452","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c36bc98d110e7bf5c5b5ea780765d705019fe82964f91d619d999e813bf2a5d3","sha256:78670bd90de9cac47f6b9cfbc5b39b5657202d102f74a2bba635a7c2712a045c"],"state_sha256":"e938e5e1cc39315c5da9d056d8161b30a375310938609e076cf76c93e3996173"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"690xRCeA3xneCNHXm+J5phMBBjcMRIzbuW72CbBunupQ+/nHrHyMSgaO7cHUE4UH1jUHDOOc8jX7AllzeXcmCQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T09:03:00.396154Z","bundle_sha256":"e8ff454bb23e81462a572abf12acc0a3f54831fb2e573f264f804614787d3bdf"}}