{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:O5GKYOZGX252BLQMTSYJ4B7673","short_pith_number":"pith:O5GKYOZG","canonical_record":{"source":{"id":"2404.03818","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-04T21:57:11Z","cross_cats_sorted":[],"title_canon_sha256":"078eae838f4ea4d462bfd7ae9b67b3be4534e99007d2960642f672c57e20ab81","abstract_canon_sha256":"eff7fb6ce67ea9aad8f0da05147fddae2225fc952fe326f82b5c3fa82d8acc9e"},"schema_version":"1.0"},"canonical_sha256":"774cac3b26bebba0ae0c9cb09e07fefedb239ceeba6c691d7a1550ad51c23b6e","source":{"kind":"arxiv","id":"2404.03818","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.03818","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"arxiv_version","alias_value":"2404.03818v4","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03818","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_12","alias_value":"O5GKYOZGX252","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_16","alias_value":"O5GKYOZGX252BLQM","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_8","alias_value":"O5GKYOZG","created_at":"2026-07-05T10:53:44Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:O5GKYOZGX252BLQMTSYJ4B7673","target":"record","payload":{"canonical_record":{"source":{"id":"2404.03818","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-04T21:57:11Z","cross_cats_sorted":[],"title_canon_sha256":"078eae838f4ea4d462bfd7ae9b67b3be4534e99007d2960642f672c57e20ab81","abstract_canon_sha256":"eff7fb6ce67ea9aad8f0da05147fddae2225fc952fe326f82b5c3fa82d8acc9e"},"schema_version":"1.0"},"canonical_sha256":"774cac3b26bebba0ae0c9cb09e07fefedb239ceeba6c691d7a1550ad51c23b6e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:44.483258Z","signature_b64":"siFcSSQyLOZAct9CyM3m718J/NujtHUk5yDZJghy8RBotlxCvKVNFe92qvBfTu8ZTf6Ms/XrnMlPKIrAbQUHBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"774cac3b26bebba0ae0c9cb09e07fefedb239ceeba6c691d7a1550ad51c23b6e","last_reissued_at":"2026-07-05T10:53:44.482781Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:44.482781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2404.03818","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":"2026-07-05T10:53:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"mJpreNWvZo73knqsJeHUK4GAheqfZYn3RaEhRn1UI8QtYHWLWGwVW4uSHU63T0uFJ5tStdZxYgVRHkLCuKGFDg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:58:47.447602Z"},"content_sha256":"d007cadf79ba4be823405c17e5de40805537c8948f40d898ba63f98636c15c4d","schema_version":"1.0","event_id":"sha256:d007cadf79ba4be823405c17e5de40805537c8948f40d898ba63f98636c15c4d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:O5GKYOZGX252BLQMTSYJ4B7673","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"PRobELM: Plausibility Ranking Evaluation for Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Andreas Vlachos, Chenxi Whitehouse, Eric Chamoun, Rami Aly, Zhangdie Yuan","submitted_at":"2024-04-04T21:57:11Z","abstract_excerpt":"This paper introduces PRobELM (Plausibility Ranking Evaluation for Language Models), a benchmark designed to assess language models' ability to discern more plausible from less plausible scenarios through their parametric knowledge. While benchmarks such as TruthfulQA emphasise factual accuracy or truthfulness, and others such as COPA explore plausible scenarios without explicitly incorporating world knowledge, PRobELM seeks to bridge this gap by evaluating models' capabilities to prioritise plausible scenarios that leverage world knowledge over less plausible alternatives. This design allows "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03818","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/2404.03818/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":"2026-07-05T10:53:44Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JeBSPCB/ZqmUqmGRcbqaksy7iJ5E8uVxfQi5MIJ+GgPZiRC9yO+W3DpvJO9MhfBux9EcK3g1d3qY9LgL1XQRCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-06T09:58:47.448114Z"},"content_sha256":"36e1f729fda58940a3b9f022861e5f3c65f4298d7d37dc7a068847ae984ee868","schema_version":"1.0","event_id":"sha256:36e1f729fda58940a3b9f022861e5f3c65f4298d7d37dc7a068847ae984ee868"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/O5GKYOZGX252BLQMTSYJ4B7673/bundle.json","state_url":"https://pith.science/pith/O5GKYOZGX252BLQMTSYJ4B7673/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/O5GKYOZGX252BLQMTSYJ4B7673/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":"2026-08-06T09:58:47Z","links":{"resolver":"https://pith.science/pith/O5GKYOZGX252BLQMTSYJ4B7673","bundle":"https://pith.science/pith/O5GKYOZGX252BLQMTSYJ4B7673/bundle.json","state":"https://pith.science/pith/O5GKYOZGX252BLQMTSYJ4B7673/state.json","well_known_bundle":"https://pith.science/.well-known/pith/O5GKYOZGX252BLQMTSYJ4B7673/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:O5GKYOZGX252BLQMTSYJ4B7673","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":"eff7fb6ce67ea9aad8f0da05147fddae2225fc952fe326f82b5c3fa82d8acc9e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-04T21:57:11Z","title_canon_sha256":"078eae838f4ea4d462bfd7ae9b67b3be4534e99007d2960642f672c57e20ab81"},"schema_version":"1.0","source":{"id":"2404.03818","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2404.03818","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"arxiv_version","alias_value":"2404.03818v4","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.03818","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_12","alias_value":"O5GKYOZGX252","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_16","alias_value":"O5GKYOZGX252BLQM","created_at":"2026-07-05T10:53:44Z"},{"alias_kind":"pith_short_8","alias_value":"O5GKYOZG","created_at":"2026-07-05T10:53:44Z"}],"graph_snapshots":[{"event_id":"sha256:36e1f729fda58940a3b9f022861e5f3c65f4298d7d37dc7a068847ae984ee868","target":"graph","created_at":"2026-07-05T10:53:44Z","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/2404.03818/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"This paper introduces PRobELM (Plausibility Ranking Evaluation for Language Models), a benchmark designed to assess language models' ability to discern more plausible from less plausible scenarios through their parametric knowledge. While benchmarks such as TruthfulQA emphasise factual accuracy or truthfulness, and others such as COPA explore plausible scenarios without explicitly incorporating world knowledge, PRobELM seeks to bridge this gap by evaluating models' capabilities to prioritise plausible scenarios that leverage world knowledge over less plausible alternatives. This design allows ","authors_text":"Andreas Vlachos, Chenxi Whitehouse, Eric Chamoun, Rami Aly, Zhangdie Yuan","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-04T21:57:11Z","title":"PRobELM: Plausibility Ranking Evaluation for Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.03818","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:d007cadf79ba4be823405c17e5de40805537c8948f40d898ba63f98636c15c4d","target":"record","created_at":"2026-07-05T10:53:44Z","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":"eff7fb6ce67ea9aad8f0da05147fddae2225fc952fe326f82b5c3fa82d8acc9e","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-04-04T21:57:11Z","title_canon_sha256":"078eae838f4ea4d462bfd7ae9b67b3be4534e99007d2960642f672c57e20ab81"},"schema_version":"1.0","source":{"id":"2404.03818","kind":"arxiv","version":4}},"canonical_sha256":"774cac3b26bebba0ae0c9cb09e07fefedb239ceeba6c691d7a1550ad51c23b6e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"774cac3b26bebba0ae0c9cb09e07fefedb239ceeba6c691d7a1550ad51c23b6e","first_computed_at":"2026-07-05T10:53:44.482781Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T10:53:44.482781Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"siFcSSQyLOZAct9CyM3m718J/NujtHUk5yDZJghy8RBotlxCvKVNFe92qvBfTu8ZTf6Ms/XrnMlPKIrAbQUHBw==","signature_status":"signed_v1","signed_at":"2026-07-05T10:53:44.483258Z","signed_message":"canonical_sha256_bytes"},"source_id":"2404.03818","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d007cadf79ba4be823405c17e5de40805537c8948f40d898ba63f98636c15c4d","sha256:36e1f729fda58940a3b9f022861e5f3c65f4298d7d37dc7a068847ae984ee868"],"state_sha256":"4a41def51a9966dce9e07185d84d914c2c1433d5566995c44bf9622852274ef4"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zA8ugvyLQyrrkHgmeSA4/Fi4RXR9aBX9q4Z6L8KpM4Ac8kvMPNFfjdpIxf+5xb9cGjAoZFZyQljFd0kL6cl4DQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-06T09:58:47.452163Z","bundle_sha256":"38d32cbaca15c2017c07faca91595ab52861147d9ebb1a8f7c3cbea4fd41bad8"}}