Guardado en:
| Autores principales: | , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | https://arxiv.org/abs/2601.05403 |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911602403442688 |
|---|---|
| author | Liu, Zhiwei Cao, Yupen Jiang, Yuechen Kabir, Mohsinul Giannouris, Polydoros Xu, Chen Xu, Ziyang Zhu, Tianlei Tariquzzaman, Md. Papadopoulos, Triantafillos Wang, Yan Qian, Lingfei Peng, Xueqing Xie, Zhuohan Yuan, Ye Almheiri, Saeed Alnajjar, Abdulrazzaq Chen, Mingbin Stuart, Harry Thompson, Paul Tiwari, Prayag Lopez-Lira, Alejandro Liu, Xue Huang, Jimin Ananiadou, Sophia |
| author_facet | Liu, Zhiwei Cao, Yupen Jiang, Yuechen Kabir, Mohsinul Giannouris, Polydoros Xu, Chen Xu, Ziyang Zhu, Tianlei Tariquzzaman, Md. Papadopoulos, Triantafillos Wang, Yan Qian, Lingfei Peng, Xueqing Xie, Zhuohan Yuan, Ye Almheiri, Saeed Alnajjar, Abdulrazzaq Chen, Mingbin Stuart, Harry Thompson, Paul Tiwari, Prayag Lopez-Lira, Alejandro Liu, Xue Huang, Jimin Ananiadou, Sophia |
| contents | Large language models (LLMs) have been widely applied across various domains of finance. Since their training data are largely derived from human-authored corpora, LLMs may inherit a range of human biases. Behavioral biases can lead to instability and uncertainty in decision-making, particularly when processing financial information. However, existing research on LLM bias has mainly focused on direct questioning or simplified, general-purpose settings, with limited consideration of the complex real-world financial environments and high-risk, context-sensitive, multilingual financial misinformation detection tasks MFMD. In this work, we propose MFMDScen, a comprehensive benchmark for evaluating behavioral biases of LLMs in MFMD across diverse economic scenarios. In collaboration with financial experts, we construct three types of complex financial scenarios: (i) role- and personality-based, (ii) role- and region-based, and (iii) role-based scenarios incorporating ethnicity and religious beliefs. We further develop a multilingual financial misinformation dataset covering English, Chinese, Greek, and Bengali. By integrating these scenarios with misinformation claims, MFMDScen enables a systematic evaluation of 22 mainstream LLMs. Our findings reveal that pronounced behavioral biases persist across both commercial and open-source models. This project is available at https://github.com/lzw108/FMD. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_05403 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Same Claim, Different Judgment: Benchmarking Scenario-Induced Bias in Multilingual Financial Misinformation Detection Liu, Zhiwei Cao, Yupen Jiang, Yuechen Kabir, Mohsinul Giannouris, Polydoros Xu, Chen Xu, Ziyang Zhu, Tianlei Tariquzzaman, Md. Papadopoulos, Triantafillos Wang, Yan Qian, Lingfei Peng, Xueqing Xie, Zhuohan Yuan, Ye Almheiri, Saeed Alnajjar, Abdulrazzaq Chen, Mingbin Stuart, Harry Thompson, Paul Tiwari, Prayag Lopez-Lira, Alejandro Liu, Xue Huang, Jimin Ananiadou, Sophia Computation and Language Large language models (LLMs) have been widely applied across various domains of finance. Since their training data are largely derived from human-authored corpora, LLMs may inherit a range of human biases. Behavioral biases can lead to instability and uncertainty in decision-making, particularly when processing financial information. However, existing research on LLM bias has mainly focused on direct questioning or simplified, general-purpose settings, with limited consideration of the complex real-world financial environments and high-risk, context-sensitive, multilingual financial misinformation detection tasks MFMD. In this work, we propose MFMDScen, a comprehensive benchmark for evaluating behavioral biases of LLMs in MFMD across diverse economic scenarios. In collaboration with financial experts, we construct three types of complex financial scenarios: (i) role- and personality-based, (ii) role- and region-based, and (iii) role-based scenarios incorporating ethnicity and religious beliefs. We further develop a multilingual financial misinformation dataset covering English, Chinese, Greek, and Bengali. By integrating these scenarios with misinformation claims, MFMDScen enables a systematic evaluation of 22 mainstream LLMs. Our findings reveal that pronounced behavioral biases persist across both commercial and open-source models. This project is available at https://github.com/lzw108/FMD. |
| title | Same Claim, Different Judgment: Benchmarking Scenario-Induced Bias in Multilingual Financial Misinformation Detection |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2601.05403 |