# FinTrace: Holistic Trajectory-Level Evaluation of LLM Tool Calling for Long-Horizon Financial Tasks

arXiv: https://arxiv.org/abs/2604.10015 · HTML: https://mlsanity.com/paper/2604.10015/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Financial trading and forecasting agents, one of 16 directions in LLM agents and tool use.
- Builds on 35 of its 44 references that are in the map.
- First posted on arXiv 11 Apr 2026.

## Placed in

- LLM agents and tool use / [Financial trading and forecasting agents](https://mlsanity.com/agents/financial-trading-and-forecasting-agents/)
- LLM alignment / RLHF / preference optimisation / [Preference alignment for recommenders and RAG](https://mlsanity.com/alignment/preference-alignment-for-recommenders-and-rag/)

## Builds on (most cited first, 30 of 35)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs](https://mlsanity.com/paper/2307.16789/) (2023)
- [MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark](https://mlsanity.com/paper/2406.01574/) (2024)
- [AgentBench: Evaluating LLMs as Agents](https://mlsanity.com/paper/2308.03688/) (2023)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [API-Bank: A Comprehensive Benchmark for Tool-Augmented LLMs](https://mlsanity.com/paper/2304.08244/) (2023)
- [TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks](https://mlsanity.com/paper/2412.14161/) (2024)
- [FinanceBench: A New Benchmark for Financial Question Answering](https://mlsanity.com/paper/2311.11944/) (2023)
- [Survey on Evaluation of LLM-based Agents](https://mlsanity.com/paper/2503.16416/) (2025)
- [Trial and Error: Exploration-Based Trajectory Optimization for LLM Agents](https://mlsanity.com/paper/2403.02502/) (2024)
- [FinBen: A Holistic Financial Benchmark for Large Language Models](https://mlsanity.com/paper/2402.12659/) (2024)
- [FinCon: A Synthesized LLM Multi-Agent System with Conceptual Verbal Reinforcement for Enhanced Financial Decision Making](https://mlsanity.com/paper/2407.06567/) (2024)
- [MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks via MCP Servers](https://mlsanity.com/paper/2508.20453/) (2025)
- [T-Eval: Evaluating the Tool Utilization Capability of Large Language Models Step by Step](https://mlsanity.com/paper/2312.14033/) (2023)
- [ToolGen: Unified Tool Retrieval and Calling via Generation](https://mlsanity.com/paper/2410.03439/) (2024)
- [INVESTORBENCH: A Benchmark for Financial Decision-Making Tasks with LLM-based Agent](https://mlsanity.com/paper/2412.18174/) (2024)
- [LLM Agents Making Agent Tools](https://mlsanity.com/paper/2502.11705/) (2025)
- [LiveMCPBench: Can Agents Navigate an Ocean of MCP Tools?](https://mlsanity.com/paper/2508.01780/) (2025)
- [Advancing Tool-Augmented Large Language Models: Integrating Insights from Errors in Inference Trees](https://mlsanity.com/paper/2406.07115/) (2024)
- [FinAgentBench: A Benchmark Dataset for Agentic Retrieval in Financial Question Answering](https://mlsanity.com/paper/2508.14052/) (2025)
- [FinSearchComp: Towards a Realistic, Expert-Level Evaluation of Financial Search and Reasoning](https://mlsanity.com/paper/2509.13160/) (2025)
- [LongFuncEval: Measuring the effectiveness of long context models for function calling](https://mlsanity.com/paper/2505.10570/) (2025)
- [MultiFinBen: Benchmarking Large Language Models for Multilingual and Multimodal Financial Application](https://mlsanity.com/paper/2506.14028/) (2025)
- [FinDeepResearch: Evaluating Deep Research Agents in Rigorous Financial Analysis](https://mlsanity.com/paper/2510.13936/) (2025)
- [The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestration](https://mlsanity.com/paper/2603.22862/) (2026)
- [Improving Tool Retrieval by Leveraging Large Language Models for Query Generation](https://mlsanity.com/paper/2412.03573/) (2024)

## Built on by (most cited first, 4 of 4)

- [DFAH-Bench: Benchmarking Observable Agent Instability in Financial Decision-Making](https://mlsanity.com/paper/2607.20491/) (2026)
- [BLINDSPOT: A Benchmark for Safety and Refusal Calibration in Long-Horizon Tool-Using Agents](https://mlsanity.com/paper/2609.16305/) (2026)
- [Fin_Agent at FinEval 2026: Task-Aware Methods for Multilingual Financial NLP](https://mlsanity.com/paper/2610.14101/) (2026)
- [Judges Follow the Reference: The Derivation Gap in Verifying Tool-Call Arguments](https://mlsanity.com/paper/2610.18497/) (2026)

## Abstract (arXiv)

Recent studies demonstrate that tool-calling capability enables large language models (LLMs) to interact with external environments for long-horizon financial tasks. While existing benchmarks have begun evaluating financial tool calling, they focus on limited scenarios and rely on call-level metrics that fail to capture trajectory-level reasoning quality. To address this gap, we introduce FinTrace, a benchmark comprising 800 expert-annotated trajectories spanning 34 real-world financial task categories across multiple difficulty levels. FinTrace employs a rubric-based evaluation protocol with nine metrics organized along four axes -- action correctness, execution efficiency, process quality, and output quality -- enabling fine-grained assessment of LLM tool-calling behavior. Our evaluation of 13 LLMs reveals that while frontier models achieve strong tool selection, all models struggle with information utilization and final answer quality, exposing a critical gap between invoking the right tools and reasoning effectively over their outputs. To move beyond diagnosis, we construct FinTrace-Training, the first trajectory-level preference dataset for financial tool-calling, containing 8,196 curated trajectories with tool-augmented contexts and preference pairs. We fine-tune Qwen-3-8B/32B using supervised fine-tuning followed by direct preference optimization (DPO) and show that training on FinTrace-Training consistently improves intermediate reasoning metrics, with DPO more effectively suppressing failure modes. However, end-to-end answer quality remains a bottleneck, indicating that trajectory-level improvements do not yet fully propagate to final output quality.