{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2025:6GK2DY56XISUWB6JVOTS4WTZRI","short_pith_number":"pith:6GK2DY56","canonical_record":{"source":{"id":"2503.05061","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T00:42:08Z","cross_cats_sorted":[],"title_canon_sha256":"2f65519053e3530a5c00ac432ec8f0546cc0356daece3150e14e8ecea29ad9a2","abstract_canon_sha256":"390e87ca7ce89df24246429ebdea2c4b8ad106bd30fec7c6219df38602184d9c"},"schema_version":"1.0"},"canonical_sha256":"f195a1e3beba254b07c9aba72e5a798a262d4cf82aad5e3126092f5074639c08","source":{"kind":"arxiv","id":"2503.05061","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.05061","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"arxiv_version","alias_value":"2503.05061v3","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.05061","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_12","alias_value":"6GK2DY56XISU","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_16","alias_value":"6GK2DY56XISUWB6J","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_8","alias_value":"6GK2DY56","created_at":"2026-08-12T01:24:11Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2025:6GK2DY56XISUWB6JVOTS4WTZRI","target":"record","payload":{"canonical_record":{"source":{"id":"2503.05061","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T00:42:08Z","cross_cats_sorted":[],"title_canon_sha256":"2f65519053e3530a5c00ac432ec8f0546cc0356daece3150e14e8ecea29ad9a2","abstract_canon_sha256":"390e87ca7ce89df24246429ebdea2c4b8ad106bd30fec7c6219df38602184d9c"},"schema_version":"1.0"},"canonical_sha256":"f195a1e3beba254b07c9aba72e5a798a262d4cf82aad5e3126092f5074639c08","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-12T01:24:11.693989Z","signature_b64":"iJh2N+VpqgqZC9amAN7N9OQQNmWECTMDgE8Esz9zyjg3uoybFedqGevk2802/V8h2mWZBaWUMdZecxTzLABuDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f195a1e3beba254b07c9aba72e5a798a262d4cf82aad5e3126092f5074639c08","last_reissued_at":"2026-08-12T01:24:11.691919Z","signature_status":"signed_v1","first_computed_at":"2026-08-12T01:24:11.691919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2503.05061","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-08-12T01:24:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"J/YSCT0mBkLS1UaONK8y7AWjC3qNXE7geM9lUczLW6wh/QfDw0AiT0/fUH2VHUJLQQmAPSLfg2E1XWwLLj09Ag==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T05:38:43.631746Z"},"content_sha256":"a17db4e5f7b3c25b8a49648da2f987a914724f2a5d64da1d2655be25d4626250","schema_version":"1.0","event_id":"sha256:a17db4e5f7b3c25b8a49648da2f987a914724f2a5d64da1d2655be25d4626250"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2025:6GK2DY56XISUWB6JVOTS4WTZRI","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Charles Lovering, Chris Tanner, Michael Krumdick, Seth Ebner, Varshini Reddy","submitted_at":"2025-03-07T00:42:08Z","abstract_excerpt":"Reliable evaluation of large language models (LLMs) is critical as their deployment rapidly expands, particularly in high-stakes domains such as business and finance. The LLM-as-a-Judge framework, which uses prompted LLMs to evaluate response quality, is appealing due to its scalability, low cost, and strong correlations with human stylistic preferences. However, it remains unclear how accurately these methods can assess response quality in domains where correctness matters more than style. To address this gap, we introduce the Business and Finance Fundamentals Benchmark (BFF-Bench), a dataset"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.05061","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/2503.05061/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":"2026-08-12T01:24:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6VrP3nDeSuUY4CDxHdJ02jLl2p0YObKWNrVxJsU7avMmB5gv7fudAWOMqc0STSoz1068lxHtwCAAHdN+kh9ICA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-17T05:38:43.632532Z"},"content_sha256":"3084b275acb8902b9495d71db9b2f2331be9063429d55a404404c066ece718e3","schema_version":"1.0","event_id":"sha256:3084b275acb8902b9495d71db9b2f2331be9063429d55a404404c066ece718e3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6GK2DY56XISUWB6JVOTS4WTZRI/bundle.json","state_url":"https://pith.science/pith/6GK2DY56XISUWB6JVOTS4WTZRI/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6GK2DY56XISUWB6JVOTS4WTZRI/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":"2026-08-17T05:38:43Z","links":{"resolver":"https://pith.science/pith/6GK2DY56XISUWB6JVOTS4WTZRI","bundle":"https://pith.science/pith/6GK2DY56XISUWB6JVOTS4WTZRI/bundle.json","state":"https://pith.science/pith/6GK2DY56XISUWB6JVOTS4WTZRI/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6GK2DY56XISUWB6JVOTS4WTZRI/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:6GK2DY56XISUWB6JVOTS4WTZRI","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":"390e87ca7ce89df24246429ebdea2c4b8ad106bd30fec7c6219df38602184d9c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T00:42:08Z","title_canon_sha256":"2f65519053e3530a5c00ac432ec8f0546cc0356daece3150e14e8ecea29ad9a2"},"schema_version":"1.0","source":{"id":"2503.05061","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2503.05061","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"arxiv_version","alias_value":"2503.05061v3","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.05061","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_12","alias_value":"6GK2DY56XISU","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_16","alias_value":"6GK2DY56XISUWB6J","created_at":"2026-08-12T01:24:11Z"},{"alias_kind":"pith_short_8","alias_value":"6GK2DY56","created_at":"2026-08-12T01:24:11Z"}],"graph_snapshots":[{"event_id":"sha256:3084b275acb8902b9495d71db9b2f2331be9063429d55a404404c066ece718e3","target":"graph","created_at":"2026-08-12T01:24:11Z","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/2503.05061/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Reliable evaluation of large language models (LLMs) is critical as their deployment rapidly expands, particularly in high-stakes domains such as business and finance. The LLM-as-a-Judge framework, which uses prompted LLMs to evaluate response quality, is appealing due to its scalability, low cost, and strong correlations with human stylistic preferences. However, it remains unclear how accurately these methods can assess response quality in domains where correctness matters more than style. To address this gap, we introduce the Business and Finance Fundamentals Benchmark (BFF-Bench), a dataset","authors_text":"Charles Lovering, Chris Tanner, Michael Krumdick, Seth Ebner, Varshini Reddy","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T00:42:08Z","title":"No Free Labels: Limitations of LLM-as-a-Judge Without Human Grounding"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.05061","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:a17db4e5f7b3c25b8a49648da2f987a914724f2a5d64da1d2655be25d4626250","target":"record","created_at":"2026-08-12T01:24:11Z","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":"390e87ca7ce89df24246429ebdea2c4b8ad106bd30fec7c6219df38602184d9c","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2025-03-07T00:42:08Z","title_canon_sha256":"2f65519053e3530a5c00ac432ec8f0546cc0356daece3150e14e8ecea29ad9a2"},"schema_version":"1.0","source":{"id":"2503.05061","kind":"arxiv","version":3}},"canonical_sha256":"f195a1e3beba254b07c9aba72e5a798a262d4cf82aad5e3126092f5074639c08","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f195a1e3beba254b07c9aba72e5a798a262d4cf82aad5e3126092f5074639c08","first_computed_at":"2026-08-12T01:24:11.691919Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-12T01:24:11.691919Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"iJh2N+VpqgqZC9amAN7N9OQQNmWECTMDgE8Esz9zyjg3uoybFedqGevk2802/V8h2mWZBaWUMdZecxTzLABuDg==","signature_status":"signed_v1","signed_at":"2026-08-12T01:24:11.693989Z","signed_message":"canonical_sha256_bytes"},"source_id":"2503.05061","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:a17db4e5f7b3c25b8a49648da2f987a914724f2a5d64da1d2655be25d4626250","sha256:3084b275acb8902b9495d71db9b2f2331be9063429d55a404404c066ece718e3"],"state_sha256":"1edc253c4434db90ca30c91c7f8086e085c2bf44df242ed76252722868042d9e"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ajMvRVmZuplr5urCS0Aiy9b+B7n/bpGVPRy7saueBWZy71ySJaTVIPr1vG3/RJ4Z7WD0Wp2kUsfShsYKDXcdBw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-17T05:38:43.639257Z","bundle_sha256":"9d1e5450d3393d3de324982a5dba96ee4c90f0cf316523b6fd13d18c37e8c2a2"}}