Research map / Multi-agent systems and games / LLM agents
Towards Professional-Grade Financial Agents: Benchmarking, Tooling, and Structured Reasoning
Cheng Huang, Jinghua Piao, Ranran Wang, Yong Li
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Introduces ProFinR, a benchmark of 528 expert-designed professional finance tasks across four domains, along with tooling and structured reasoning for LLM financial agents.
Problem
LLM agents lack holistic benchmarks for realistic professional financial workflows, and unstructured reasoning in them is fragile.
Approach
A holistic professional-finance benchmark (ProFinR) with expert-designed tasks, paired with tooling and structured reasoning to make financial agents more robust than unstructured reasoning.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| wrap LLM with ProFinAgent (tools, DAG planner, case memory) | GPT-5.2 · ProFinR | Total score delta vs native: +49.81% | Improved |
| wrap LLM with ProFinAgent | Qwen3-4B-Instruct-2507 · ProFinR | Total score delta vs native: +27.84% | Improved |
| add DAG parallel execution (ablation) | ProFinAgent on DeepSeek-R1 · ProFinR | total execution time: 42,579.9 s -> 22,547.2 s (47.1% reduction) | Improved |
| add 53-tool library to ReAct without structure (ablation) | DeepSeek-R1 ReAct · ProFinR L3 | accuracy: 29.45% -> 21.70% | Hurt |
| add tool use to ReAct (ablation) | Qwen3-32B · ProFinR L1 | accuracy: 5.28% -> 40.38% | Improved |
| add DAG planner (ablation) | DeepSeek-R1 ReAct · ProFinR L3 | accuracy: 29.45% -> 33.33% | Improved |
| add case-based memory (ablation) | DeepSeek-R1 · ProFinR L1 | accuracy: 43.75% -> 53.36% | Improved |
| combine DAG and CBM versus either alone (ablation) | DeepSeek-R1 · ProFinR L3 | accuracy: 33.33% / 31.78% -> 54.26% | Improved |
| domain-specific financial fine-tuning (Fin-R1) | native inference · ProFinR | total score: 1.70% (Fin-R1) vs 1.70% (Qwen3-4B) | No clear effect |
| use ProFinAgent instead of tree-search tool planners | GPT-4o · RestBench-TMDB | avg: 74.50% -> 81.25% | Improved |
| ProFinAgent vs ToolTree win rate | GPT-4o-mini · RestBench-Spotify | win rate: 72.18% -> 71.93% | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TOOLTREE — ToolTree: Efficient LLM Tool Planning via Dual-Feedback Monte Carlo Tree Search and Bidirectional Pruning (ICLR 2026) · better on 3 of 3 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Experimental setup
Datasets: ProFinR
Benchmarks: ProFinR
Closest papers by meaning
- EnterpriseBench: Evaluating LLM Agents on End-to-End Spreadsheet Tasks in Finance (NeurIPS 2026 · similarity 0.88)
- BankerToolBench: Evaluating AI Agents in End-to-End Investment Banking Workflows (NeurIPS 2026 · similarity 0.88)
- FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning (ICLR 2026 · similarity 0.88)
- FinMMDocR: Benchmarking Financial Multimodal Reasoning with Scenario Awareness, Document Understanding, and Multi-Step Computation (AAAI 2026 · similarity 0.86)
- \$OneMillion-Bench: How Far are Language Agents from Human Experts? (NeurIPS 2026 · similarity 0.86)
- FinBen: A Holistic Financial Benchmark for Large Language Models (NeurIPS 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
