Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866910227547291648 |
|---|---|
| author | Wang, Yan Han, Yi Qian, Lingfei He, Yueru Peng, Xueqing Feng, Dongji Xie, Zhuohan Zhang, Vincent Jim Guo, Rosie Mo, Fengran Huang, Jimin Chen, Yankai Liu, Xue Nie, Jian-Yun |
| author_facet | Wang, Yan Han, Yi Qian, Lingfei He, Yueru Peng, Xueqing Feng, Dongji Xie, Zhuohan Zhang, Vincent Jim Guo, Rosie Mo, Fengran Huang, Jimin Chen, Yankai Liu, Xue Nie, Jian-Yun |
| contents | Most recommendation benchmarks evaluate how well a model imitates user behavior. In financial advisory, however, observed actions can be noisy or short-sighted under market volatility and may conflict with a user's long-term goals. Treating what users chose as the sole ground truth, therefore, conflates behavioral imitation with decision quality. We introduce Conv-FinRe, a conversational and longitudinal benchmark for stock recommendation that evaluates LLMs beyond behavior matching. Given an onboarding interview, step-wise market context, and advisory dialogues, models must generate rankings over a fixed investment horizon. Crucially, Conv-FinRe provides multi-view references that distinguish descriptive behavior from normative utility grounded in investor-specific risk preferences, enabling diagnosis of whether an LLM follows rational analysis, mimics user noise, or is driven by market momentum. We build the benchmark from real market data and human decision trajectories, instantiate controlled advisory conversations, and evaluate a suite of state-of-the-art LLMs. Results reveal a persistent tension between rational decision quality and behavioral alignment: models that perform well on utility-based ranking often fail to match user choices, whereas behaviorally aligned models can overfit short-term noise. The dataset is publicly released on Hugging Face, and the codebase is available on GitHub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_16990 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation Wang, Yan Han, Yi Qian, Lingfei He, Yueru Peng, Xueqing Feng, Dongji Xie, Zhuohan Zhang, Vincent Jim Guo, Rosie Mo, Fengran Huang, Jimin Chen, Yankai Liu, Xue Nie, Jian-Yun Artificial Intelligence Computational Engineering, Finance, and Science Most recommendation benchmarks evaluate how well a model imitates user behavior. In financial advisory, however, observed actions can be noisy or short-sighted under market volatility and may conflict with a user's long-term goals. Treating what users chose as the sole ground truth, therefore, conflates behavioral imitation with decision quality. We introduce Conv-FinRe, a conversational and longitudinal benchmark for stock recommendation that evaluates LLMs beyond behavior matching. Given an onboarding interview, step-wise market context, and advisory dialogues, models must generate rankings over a fixed investment horizon. Crucially, Conv-FinRe provides multi-view references that distinguish descriptive behavior from normative utility grounded in investor-specific risk preferences, enabling diagnosis of whether an LLM follows rational analysis, mimics user noise, or is driven by market momentum. We build the benchmark from real market data and human decision trajectories, instantiate controlled advisory conversations, and evaluate a suite of state-of-the-art LLMs. Results reveal a persistent tension between rational decision quality and behavioral alignment: models that perform well on utility-based ranking often fail to match user choices, whereas behaviorally aligned models can overfit short-term noise. The dataset is publicly released on Hugging Face, and the codebase is available on GitHub. |
| title | Conv-FinRe: A Conversational and Longitudinal Benchmark for Utility-Grounded Financial Recommendation |
| topic | Artificial Intelligence Computational Engineering, Finance, and Science |
| url | https://arxiv.org/abs/2602.16990 |