{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:S6QZIOVA7ZI24EONBW4676SPM3","short_pith_number":"pith:S6QZIOVA","schema_version":"1.0","canonical_sha256":"97a1943aa0fe51ae11cd0db9effa4f66da36b78f57444e09382cbdb205ba3c20","source":{"kind":"arxiv","id":"2510.18941","version":2},"attestation_state":"computed","paper":{"title":"ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ellie Evans, Jaehun Jung, Jan Kautz, Jiaqi Zeng, Pavlo Molchanov, Shizhe Diao, Ximing Lu, Yejin Choi, Yi Dong, Zhilin Wang","submitted_at":"2025-10-21T17:59:44Z","abstract_excerpt":"Evaluating progress in large language models (LLMs) is often constrained by the challenge of verifying responses, limiting assessments to tasks like mathematics, programming, and short-form question-answering. However, many real-world applications require evaluating LLMs in processing professional documents, synthesizing information, and generating comprehensive reports in response to user queries. We introduce ProfBench: a set of over 7000 response-criterion pairs as evaluated by human-experts with professional knowledge across Physics PhD, Chemistry PhD, Finance MBA and Consulting MBA. We bu"},"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":"2510.18941","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2025-10-21T17:59:44Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"d13109efdd6d5fffa06672505a60fcdfe2abe6a897e49eb2ce870651ed66197f","abstract_canon_sha256":"2be5f3e5e01698ba747edd330f7a2409a7096df4f9a0e6bea683d76d9430becf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:02:57.683812Z","signature_b64":"lhiXLLpNy5n1pKPKQblqApWn/jYHdDkb07qcaSwcbDHAhTwlTJRvk0L8iawsKBpxVbhOHGuFZEXgTFq8uhqlDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"97a1943aa0fe51ae11cd0db9effa4f66da36b78f57444e09382cbdb205ba3c20","last_reissued_at":"2026-05-20T00:02:57.682917Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:02:57.682917Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ellie Evans, Jaehun Jung, Jan Kautz, Jiaqi Zeng, Pavlo Molchanov, Shizhe Diao, Ximing Lu, Yejin Choi, Yi Dong, Zhilin Wang","submitted_at":"2025-10-21T17:59:44Z","abstract_excerpt":"Evaluating progress in large language models (LLMs) is often constrained by the challenge of verifying responses, limiting assessments to tasks like mathematics, programming, and short-form question-answering. However, many real-world applications require evaluating LLMs in processing professional documents, synthesizing information, and generating comprehensive reports in response to user queries. We introduce ProfBench: a set of over 7000 response-criterion pairs as evaluated by human-experts with professional knowledge across Physics PhD, Chemistry PhD, Finance MBA and Consulting MBA. We bu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.18941","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/2510.18941/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":"2510.18941","created_at":"2026-05-20T00:02:57.683072+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.18941v2","created_at":"2026-05-20T00:02:57.683072+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.18941","created_at":"2026-05-20T00:02:57.683072+00:00"},{"alias_kind":"pith_short_12","alias_value":"S6QZIOVA7ZI2","created_at":"2026-05-20T00:02:57.683072+00:00"},{"alias_kind":"pith_short_16","alias_value":"S6QZIOVA7ZI24EON","created_at":"2026-05-20T00:02:57.683072+00:00"},{"alias_kind":"pith_short_8","alias_value":"S6QZIOVA","created_at":"2026-05-20T00:02:57.683072+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"2605.17554","citing_title":"Evaluating Deep Research Agents on Expert Consulting Work: A Benchmark with Verifiers, Rubrics, and Cognitive Traps","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2605.18630","citing_title":"SCICONVBENCH: Benchmarking LLMs on Multi-Turn Clarification for Task Formulation in Computational Science","ref_index":69,"is_internal_anchor":true},{"citing_arxiv_id":"2605.12474","citing_title":"Reward Hacking in Rubric-Based Reinforcement Learning","ref_index":32,"is_internal_anchor":true},{"citing_arxiv_id":"2604.13029","citing_title":"Visual Preference Optimization with Rubric Rewards","ref_index":45,"is_internal_anchor":true},{"citing_arxiv_id":"2604.11304","citing_title":"BankerToolBench: Evaluating AI Agents in End-to-End Investment Banking Workflows","ref_index":17,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3","json":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3.json","graph_json":"https://pith.science/api/pith-number/S6QZIOVA7ZI24EONBW4676SPM3/graph.json","events_json":"https://pith.science/api/pith-number/S6QZIOVA7ZI24EONBW4676SPM3/events.json","paper":"https://pith.science/paper/S6QZIOVA"},"agent_actions":{"view_html":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3","download_json":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3.json","view_paper":"https://pith.science/paper/S6QZIOVA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.18941&json=true","fetch_graph":"https://pith.science/api/pith-number/S6QZIOVA7ZI24EONBW4676SPM3/graph.json","fetch_events":"https://pith.science/api/pith-number/S6QZIOVA7ZI24EONBW4676SPM3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3/action/storage_attestation","attest_author":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3/action/author_attestation","sign_citation":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3/action/citation_signature","submit_replication":"https://pith.science/pith/S6QZIOVA7ZI24EONBW4676SPM3/action/replication_record"}},"created_at":"2026-05-20T00:02:57.683072+00:00","updated_at":"2026-05-20T00:02:57.683072+00:00"}