A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasoners
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866916422619234304 |
|---|---|
| author | Jiang, Bowen Xie, Yangxinyu Hao, Zhuoqun Wang, Xiaomeng Mallick, Tanwi Su, Weijie J. Taylor, Camillo J. Roth, Dan |
| author_facet | Jiang, Bowen Xie, Yangxinyu Hao, Zhuoqun Wang, Xiaomeng Mallick, Tanwi Su, Weijie J. Taylor, Camillo J. Roth, Dan |
| contents | This study introduces a hypothesis-testing framework to assess whether large language models (LLMs) possess genuine reasoning abilities or primarily depend on token bias. We go beyond evaluating LLMs on accuracy; rather, we aim to investigate their token bias in solving logical reasoning tasks. Specifically, we develop carefully controlled synthetic datasets, featuring conjunction fallacy and syllogistic problems. Our framework outlines a list of hypotheses where token biases are readily identifiable, with all null hypotheses assuming genuine reasoning capabilities of LLMs. The findings in this study suggest, with statistical guarantee, that most LLMs still struggle with logical reasoning. While they may perform well on classic problems, their success largely depends on recognizing superficial patterns with strong token bias, thereby raising concerns about their actual reasoning and generalization abilities. Codes and data are open-sourced at https://github.com/bowen-upenn/llm_token_bias. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2406_11050 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasoners Jiang, Bowen Xie, Yangxinyu Hao, Zhuoqun Wang, Xiaomeng Mallick, Tanwi Su, Weijie J. Taylor, Camillo J. Roth, Dan Computation and Language Artificial Intelligence This study introduces a hypothesis-testing framework to assess whether large language models (LLMs) possess genuine reasoning abilities or primarily depend on token bias. We go beyond evaluating LLMs on accuracy; rather, we aim to investigate their token bias in solving logical reasoning tasks. Specifically, we develop carefully controlled synthetic datasets, featuring conjunction fallacy and syllogistic problems. Our framework outlines a list of hypotheses where token biases are readily identifiable, with all null hypotheses assuming genuine reasoning capabilities of LLMs. The findings in this study suggest, with statistical guarantee, that most LLMs still struggle with logical reasoning. While they may perform well on classic problems, their success largely depends on recognizing superficial patterns with strong token bias, thereby raising concerns about their actual reasoning and generalization abilities. Codes and data are open-sourced at https://github.com/bowen-upenn/llm_token_bias. |
| title | A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasoners |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2406.11050 |