A symbolic benchmark for verifiable chain-of-thought financial reasoning. Includes executable templates, 58 topics across 12 domains, and ChainEval metrics.
-
Updated
Sep 17, 2026 - Python
A symbolic benchmark for verifiable chain-of-thought financial reasoning. Includes executable templates, 58 topics across 12 domains, and ChainEval metrics.
语义门控金融推理与资产配置平台(Alpha-R1 范式:语义门控 + 两阶段强化学习)
When Verified Financial Labels Fail: source-traced audits of released financial labels, rounding errors and paired grading consequences.
Sanitized public companion, evaluation summary, and technical report for FinVector-Market-4B
Self-improving financial-QA agent on FinQA: a governed loop that refines its own prompt + reasoning level over 3 cycles (no retraining), with a validation-guarded gate that verifies every change on held-out data.
Explainable financial sentiment reasoning engine using concept-level extraction, symbolic categories, interaction rules, and interpretable market signals.
Benchmark comparing Gemini 3 vs Fin-o1-14B on financial reasoning, with an LLM-as-judge scoring pipeline and a Next.js results dashboard
1,650 individually authored Korean office tasks for LLM evaluation and document understanding. Exact-match tax, payroll and VAT reasoning. 한국어 문서 이해와 연말정산, 급여 및 부가세 실무 데이터셋.
FinReason Cup 2026 (IEEE Big Data Cup): agentic financial reasoning, hedging & audit
To associate your repository with the financial-reasoning topic, visit your repo's landing page and select "manage topics."