{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:JLHWWKEHI3PS33VUPQWAC3OLNG","short_pith_number":"pith:JLHWWKEH","canonical_record":{"source":{"id":"2406.04145","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:10:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"18cbdbfa4be69df17511df688796c82ae6e721349e7c5e3647c287ee3e143b21","abstract_canon_sha256":"e59716a709204b2eecd743f4580afc3b46fc23e237a2739cb4c7fb4ddcc047fb"},"schema_version":"1.0"},"canonical_sha256":"4acf6b288746df2deeb47c2c016dcb69883433c705f04b231d811fb0ad52000e","source":{"kind":"arxiv","id":"2406.04145","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04145","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04145v1","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04145","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_12","alias_value":"JLHWWKEHI3PS","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_16","alias_value":"JLHWWKEHI3PS33VU","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_8","alias_value":"JLHWWKEH","created_at":"2026-07-05T08:28:21Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:JLHWWKEHI3PS33VUPQWAC3OLNG","target":"record","payload":{"canonical_record":{"source":{"id":"2406.04145","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:10:27Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"18cbdbfa4be69df17511df688796c82ae6e721349e7c5e3647c287ee3e143b21","abstract_canon_sha256":"e59716a709204b2eecd743f4580afc3b46fc23e237a2739cb4c7fb4ddcc047fb"},"schema_version":"1.0"},"canonical_sha256":"4acf6b288746df2deeb47c2c016dcb69883433c705f04b231d811fb0ad52000e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:28:21.316743Z","signature_b64":"ykMSAH3iQAr2SJw+KIPfdMKUyis+nWqVBRKI5+tGMu280/3zxPZV6VblHEva/zf18WxpGByVWu3NnvwZsZBrBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4acf6b288746df2deeb47c2c016dcb69883433c705f04b231d811fb0ad52000e","last_reissued_at":"2026-07-05T08:28:21.316332Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:28:21.316332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2406.04145","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":"2026-07-05T08:28:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"e80L1PaG14RtZRcgmCwM6Vy2C8KOp8mYAlWyWgKagxBCYKVluhrWGQGkz7fKjnFS6l1QMfl17NZIEDcfgsuwBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:53:12.731945Z"},"content_sha256":"83463ebd61be5b4363ed9ace7804418f310100959b5e80404d0a9d2349e221a2","schema_version":"1.0","event_id":"sha256:83463ebd61be5b4363ed9ace7804418f310100959b5e80404d0a9d2349e221a2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:JLHWWKEHI3PS33VUPQWAC3OLNG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Every Answer Matters: Evaluating Commonsense with Probabilistic Measures","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Andrew McCallum, Michael Boratko, Nalini Singh, Pranay Kumar Yelugam, Qi Cheng, Tim O'Gorman, Xiang Lorraine Li","submitted_at":"2024-06-06T15:10:27Z","abstract_excerpt":"Large language models have demonstrated impressive performance on commonsense tasks; however, these tasks are often posed as multiple-choice questions, allowing models to exploit systematic biases. Commonsense is also inherently probabilistic with multiple correct answers. The purpose of \"boiling water\" could be making tea and cooking, but it also could be killing germs. Existing tasks do not capture the probabilistic nature of common sense. To this end, we present commonsense frame completion (CFC), a new generative task that evaluates common sense via multiple open-ended generations. We also"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04145","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/2406.04145/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":"2026-07-05T08:28:21Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"SBigQZZPhjfvgy3N3FLV8/ESADDn/Lg5BBxamlIkf3+KFFkLVfEvHRKwYxE2Hu+2M159C701IHyedaU8eFUIDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-04T16:53:12.732451Z"},"content_sha256":"b9942e0d0c9c8875ff2d16190ca2126cffca489e157633f5a06e80e0be7481db","schema_version":"1.0","event_id":"sha256:b9942e0d0c9c8875ff2d16190ca2126cffca489e157633f5a06e80e0be7481db"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/bundle.json","state_url":"https://pith.science/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/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":"2026-08-04T16:53:12Z","links":{"resolver":"https://pith.science/pith/JLHWWKEHI3PS33VUPQWAC3OLNG","bundle":"https://pith.science/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/bundle.json","state":"https://pith.science/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/JLHWWKEHI3PS33VUPQWAC3OLNG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:JLHWWKEHI3PS33VUPQWAC3OLNG","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":"e59716a709204b2eecd743f4580afc3b46fc23e237a2739cb4c7fb4ddcc047fb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:10:27Z","title_canon_sha256":"18cbdbfa4be69df17511df688796c82ae6e721349e7c5e3647c287ee3e143b21"},"schema_version":"1.0","source":{"id":"2406.04145","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2406.04145","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"arxiv_version","alias_value":"2406.04145v1","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.04145","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_12","alias_value":"JLHWWKEHI3PS","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_16","alias_value":"JLHWWKEHI3PS33VU","created_at":"2026-07-05T08:28:21Z"},{"alias_kind":"pith_short_8","alias_value":"JLHWWKEH","created_at":"2026-07-05T08:28:21Z"}],"graph_snapshots":[{"event_id":"sha256:b9942e0d0c9c8875ff2d16190ca2126cffca489e157633f5a06e80e0be7481db","target":"graph","created_at":"2026-07-05T08:28:21Z","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/2406.04145/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large language models have demonstrated impressive performance on commonsense tasks; however, these tasks are often posed as multiple-choice questions, allowing models to exploit systematic biases. Commonsense is also inherently probabilistic with multiple correct answers. The purpose of \"boiling water\" could be making tea and cooking, but it also could be killing germs. Existing tasks do not capture the probabilistic nature of common sense. To this end, we present commonsense frame completion (CFC), a new generative task that evaluates common sense via multiple open-ended generations. We also","authors_text":"Andrew McCallum, Michael Boratko, Nalini Singh, Pranay Kumar Yelugam, Qi Cheng, Tim O'Gorman, Xiang Lorraine Li","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:10:27Z","title":"Every Answer Matters: Evaluating Commonsense with Probabilistic Measures"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.04145","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":"sha256:83463ebd61be5b4363ed9ace7804418f310100959b5e80404d0a9d2349e221a2","target":"record","created_at":"2026-07-05T08:28:21Z","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":"e59716a709204b2eecd743f4580afc3b46fc23e237a2739cb4c7fb4ddcc047fb","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-06T15:10:27Z","title_canon_sha256":"18cbdbfa4be69df17511df688796c82ae6e721349e7c5e3647c287ee3e143b21"},"schema_version":"1.0","source":{"id":"2406.04145","kind":"arxiv","version":1}},"canonical_sha256":"4acf6b288746df2deeb47c2c016dcb69883433c705f04b231d811fb0ad52000e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"4acf6b288746df2deeb47c2c016dcb69883433c705f04b231d811fb0ad52000e","first_computed_at":"2026-07-05T08:28:21.316332Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:28:21.316332Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"ykMSAH3iQAr2SJw+KIPfdMKUyis+nWqVBRKI5+tGMu280/3zxPZV6VblHEva/zf18WxpGByVWu3NnvwZsZBrBw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:28:21.316743Z","signed_message":"canonical_sha256_bytes"},"source_id":"2406.04145","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:83463ebd61be5b4363ed9ace7804418f310100959b5e80404d0a9d2349e221a2","sha256:b9942e0d0c9c8875ff2d16190ca2126cffca489e157633f5a06e80e0be7481db"],"state_sha256":"127a6861f20b870d438cab537d29b3adeb6c1478245e716620fe0087a96c5f49"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"FZ5g/EhVCjKGvG/8UwaVZSZMK4fL5iabsV9i96d5sj+bHSab1/Gg+wyan2TMfZfN/QTsCRCDC7r9UvRciNK5Bg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-04T16:53:12.737211Z","bundle_sha256":"d5786fcdc153464d32092e497725bf326e4fdad8245c6880a217d6b72d1dcef7"}}