{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:OKHESP357JTS5B57VHI6UCBTH2","short_pith_number":"pith:OKHESP35","schema_version":"1.0","canonical_sha256":"728e493f7dfa672e87bfa9d1ea08333eb7e2a381963bb11b9e78da9b906ec517","source":{"kind":"arxiv","id":"2508.19202","version":3},"attestation_state":"computed","paper":{"title":"Demystifying Scientific Problem-Solving in LLMs by Probing Knowledge and Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alan Li, Arman Cohan, Arpan Sarkar, Doug Downey, Yixin Liu","submitted_at":"2025-08-26T17:04:23Z","abstract_excerpt":"Scientific problem solving poses unique challenges for LLMs, requiring both deep domain knowledge and the ability to apply such knowledge through complex reasoning. While automated scientific reasoners hold great promise for assisting human scientists, there is currently no widely adopted holistic benchmark for evaluating scientific reasoning, and few approaches systematically disentangle the distinct roles of knowledge and reasoning in these tasks.\n  To address these gaps, we introduce SciReas, a diverse suite of existing benchmarks for scientific reasoning tasks, and SciReas-Pro, a selective"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2508.19202","kind":"arxiv","version":3},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-08-26T17:04:23Z","cross_cats_sorted":[],"title_canon_sha256":"b80bc67c3a88dfb962f7c2cc66a7085dd9357b111820f0919c5838a702319c8d","abstract_canon_sha256":"0152514615dbf39518cfbc5f9282fb1d0c291bf4051c9064c64cd54cd80ce745"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-29T01:04:55.393619Z","signature_b64":"9uUFp4wqaMu7yJYBXsFpR+arERlQCwYmEEUhJ0Mr+0uRkXI3bb3QbknpFjBKuXaFv7ZzPo9AUjr5StuuODnNDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"728e493f7dfa672e87bfa9d1ea08333eb7e2a381963bb11b9e78da9b906ec517","last_reissued_at":"2026-05-29T01:04:55.393133Z","signature_status":"signed_v1","first_computed_at":"2026-05-29T01:04:55.393133Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Demystifying Scientific Problem-Solving in LLMs by Probing Knowledge and Reasoning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alan Li, Arman Cohan, Arpan Sarkar, Doug Downey, Yixin Liu","submitted_at":"2025-08-26T17:04:23Z","abstract_excerpt":"Scientific problem solving poses unique challenges for LLMs, requiring both deep domain knowledge and the ability to apply such knowledge through complex reasoning. While automated scientific reasoners hold great promise for assisting human scientists, there is currently no widely adopted holistic benchmark for evaluating scientific reasoning, and few approaches systematically disentangle the distinct roles of knowledge and reasoning in these tasks.\n  To address these gaps, we introduce SciReas, a diverse suite of existing benchmarks for scientific reasoning tasks, and SciReas-Pro, a selective"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.19202","kind":"arxiv","version":3},"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/2508.19202/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2508.19202","created_at":"2026-05-29T01:04:55.393190+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.19202v3","created_at":"2026-05-29T01:04:55.393190+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.19202","created_at":"2026-05-29T01:04:55.393190+00:00"},{"alias_kind":"pith_short_12","alias_value":"OKHESP357JTS","created_at":"2026-05-29T01:04:55.393190+00:00"},{"alias_kind":"pith_short_16","alias_value":"OKHESP357JTS5B57","created_at":"2026-05-29T01:04:55.393190+00:00"},{"alias_kind":"pith_short_8","alias_value":"OKHESP35","created_at":"2026-05-29T01:04:55.393190+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2","json":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2.json","graph_json":"https://pith.science/api/pith-number/OKHESP357JTS5B57VHI6UCBTH2/graph.json","events_json":"https://pith.science/api/pith-number/OKHESP357JTS5B57VHI6UCBTH2/events.json","paper":"https://pith.science/paper/OKHESP35"},"agent_actions":{"view_html":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2","download_json":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2.json","view_paper":"https://pith.science/paper/OKHESP35","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.19202&json=true","fetch_graph":"https://pith.science/api/pith-number/OKHESP357JTS5B57VHI6UCBTH2/graph.json","fetch_events":"https://pith.science/api/pith-number/OKHESP357JTS5B57VHI6UCBTH2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2/action/storage_attestation","attest_author":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2/action/author_attestation","sign_citation":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2/action/citation_signature","submit_replication":"https://pith.science/pith/OKHESP357JTS5B57VHI6UCBTH2/action/replication_record"}},"created_at":"2026-05-29T01:04:55.393190+00:00","updated_at":"2026-05-29T01:04:55.393190+00:00"}