{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:2ZU4CENAEK5M2NYRHTHTGLECZ6","short_pith_number":"pith:2ZU4CENA","schema_version":"1.0","canonical_sha256":"d669c111a022bacd37113ccf332c82cfb2a985dc2dc18df9d8f09f3d4b219ec8","source":{"kind":"arxiv","id":"2601.06401","version":2},"attestation_state":"computed","paper":{"title":"BizFinBench.v2: Towards Reliable LLMs in Finance via Real-User Data and Offline/Online Bilingual Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guilong Lu, Liwen Zhang, Rongjunchen Zhang, Shuai Jia, Xin Guo, Xuntao Guo, Zhi Yang","submitted_at":"2026-01-10T02:51:53Z","abstract_excerpt":"Large language models are becoming increasingly significant in financial applications. Nevertheless, prevailing benchmarks are largely dependent on simulated or generic data, which leads to a significant gap between reported performance and actual efficacy in real-world scenarios. To tackle this challenge, we present BizFinBench.v2, the first integrated offline and online benchmark built upon authentic user query-response data from both Chinese and U.S. equity markets. It comprises 28,860 questions across eight offline and two online tasks. Experimental results show that GPT-5 achieves a mere "},"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":"2601.06401","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-01-10T02:51:53Z","cross_cats_sorted":["cs.CL"],"title_canon_sha256":"8a4347e3769239666bd60e74d915aa76132b09be0eb024f5f78479c2cc291a9f","abstract_canon_sha256":"446d53f42c45112888df2a098ba4c3f7bd907bd7943b962d077cd27074c384dd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-14T01:22:03.084295Z","signature_b64":"ffC8C2clq9LRZluhNHO0F1mAhfkUGhcQWZBSyWwp5g5Q0CqAML3xcHpSE2fR16ACjsy/1VJfyXGHnCjZwO75Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d669c111a022bacd37113ccf332c82cfb2a985dc2dc18df9d8f09f3d4b219ec8","last_reissued_at":"2026-07-14T01:22:03.083297Z","signature_status":"signed_v1","first_computed_at":"2026-07-14T01:22:03.083297Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"BizFinBench.v2: Towards Reliable LLMs in Finance via Real-User Data and Offline/Online Bilingual Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL"],"primary_cat":"cs.AI","authors_text":"Guilong Lu, Liwen Zhang, Rongjunchen Zhang, Shuai Jia, Xin Guo, Xuntao Guo, Zhi Yang","submitted_at":"2026-01-10T02:51:53Z","abstract_excerpt":"Large language models are becoming increasingly significant in financial applications. Nevertheless, prevailing benchmarks are largely dependent on simulated or generic data, which leads to a significant gap between reported performance and actual efficacy in real-world scenarios. To tackle this challenge, we present BizFinBench.v2, the first integrated offline and online benchmark built upon authentic user query-response data from both Chinese and U.S. equity markets. It comprises 28,860 questions across eight offline and two online tasks. Experimental results show that GPT-5 achieves a mere "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.06401","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/2601.06401/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":"2601.06401","created_at":"2026-07-14T01:22:03.083755+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.06401v2","created_at":"2026-07-14T01:22:03.083755+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.06401","created_at":"2026-07-14T01:22:03.083755+00:00"},{"alias_kind":"pith_short_12","alias_value":"2ZU4CENAEK5M","created_at":"2026-07-14T01:22:03.083755+00:00"},{"alias_kind":"pith_short_16","alias_value":"2ZU4CENAEK5M2NYR","created_at":"2026-07-14T01:22:03.083755+00:00"},{"alias_kind":"pith_short_8","alias_value":"2ZU4CENA","created_at":"2026-07-14T01:22:03.083755+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2604.09418","citing_title":"Automated Instruction Revision (AIR): A Structured Comparison of Task Adaptation Strategies for LLM","ref_index":18,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6","json":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6.json","graph_json":"https://pith.science/api/pith-number/2ZU4CENAEK5M2NYRHTHTGLECZ6/graph.json","events_json":"https://pith.science/api/pith-number/2ZU4CENAEK5M2NYRHTHTGLECZ6/events.json","paper":"https://pith.science/paper/2ZU4CENA"},"agent_actions":{"view_html":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6","download_json":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6.json","view_paper":"https://pith.science/paper/2ZU4CENA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.06401&json=true","fetch_graph":"https://pith.science/api/pith-number/2ZU4CENAEK5M2NYRHTHTGLECZ6/graph.json","fetch_events":"https://pith.science/api/pith-number/2ZU4CENAEK5M2NYRHTHTGLECZ6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6/action/storage_attestation","attest_author":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6/action/author_attestation","sign_citation":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6/action/citation_signature","submit_replication":"https://pith.science/pith/2ZU4CENAEK5M2NYRHTHTGLECZ6/action/replication_record"}},"created_at":"2026-07-14T01:22:03.083755+00:00","updated_at":"2026-07-14T01:22:03.083755+00:00"}