{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:S4R4LEBDP3MTXGYR3V2I5RQ5XG","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":"1f091ede868e7c28f3cd2f17d36357dd634f71315f004008018cff5fc0437903","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-07T04:15:17Z","title_canon_sha256":"038102002f72e6884a4875171f7db819deabc69039c7d9e486e5358b9916ddba"},"schema_version":"1.0","source":{"id":"2309.06578","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2309.06578","created_at":"2026-07-05T08:00:35Z"},{"alias_kind":"arxiv_version","alias_value":"2309.06578v3","created_at":"2026-07-05T08:00:35Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2309.06578","created_at":"2026-07-05T08:00:35Z"},{"alias_kind":"pith_short_12","alias_value":"S4R4LEBDP3MT","created_at":"2026-07-05T08:00:35Z"},{"alias_kind":"pith_short_16","alias_value":"S4R4LEBDP3MTXGYR","created_at":"2026-07-05T08:00:35Z"},{"alias_kind":"pith_short_8","alias_value":"S4R4LEBD","created_at":"2026-07-05T08:00:35Z"}],"graph_snapshots":[{"event_id":"sha256:1d44cc24499359675b4bfad811c2296bcc6391e03c778b1a1a108fab90fb5b39","target":"graph","created_at":"2026-07-05T08:00:35Z","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/2309.06578/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Hypothesis formulation and testing are central to empirical research. A strong hypothesis is a best guess based on existing evidence and informed by a comprehensive view of relevant literature. However, with exponential increase in the number of scientific articles published annually, manual aggregation and synthesis of evidence related to a given hypothesis is a challenge. Our work explores the ability of current large language models (LLMs) to discern evidence in support or refute of specific hypotheses based on the text of scientific abstracts. We share a novel dataset for the task of scien","authors_text":"Jian Wu, Sai Koneru, Sarah Rajtmajer","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-07T04:15:17Z","title":"Can Large Language Models Discern Evidence for Scientific Hypotheses? Case Studies in the Social Sciences"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2309.06578","kind":"arxiv","version":3},"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":"sha256:8bdb4a06d55e148a2f9565f00d2bda85639cfb84bedfd15bcf0a962300a5c8b4","target":"record","created_at":"2026-07-05T08:00:35Z","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":"1f091ede868e7c28f3cd2f17d36357dd634f71315f004008018cff5fc0437903","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-09-07T04:15:17Z","title_canon_sha256":"038102002f72e6884a4875171f7db819deabc69039c7d9e486e5358b9916ddba"},"schema_version":"1.0","source":{"id":"2309.06578","kind":"arxiv","version":3}},"canonical_sha256":"9723c590237ed93b9b11dd748ec61db9b1e9cde2e854323c79819545a1050ff3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"9723c590237ed93b9b11dd748ec61db9b1e9cde2e854323c79819545a1050ff3","first_computed_at":"2026-07-05T08:00:35.683125Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:00:35.683125Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"+Mbolt4dyn9L0Z4QZHNghd6gIBoQzswLo0JCWVDcdkwJoXceiwFH3TyBL9U12Vl/sy3XD/JWNoP0TphcLQx/Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:00:35.683665Z","signed_message":"canonical_sha256_bytes"},"source_id":"2309.06578","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8bdb4a06d55e148a2f9565f00d2bda85639cfb84bedfd15bcf0a962300a5c8b4","sha256:1d44cc24499359675b4bfad811c2296bcc6391e03c778b1a1a108fab90fb5b39"],"state_sha256":"159cad72ab1f84b3943b087d006223e8d93ced7053bc763cd2e0aae84f2bed6a"}