MeTHanol: Modularized Thinking Language Models with Intermediate Layer Thinking, Decoding and Bootstrapping Reasoning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909000283455488 |
|---|---|
| author | Xi, Ningyuan Wang, Xiaoyu Wu, Yetao Chen, Teng Gu, Qingqing Zhao, Yue Qu, Jinxian Jiang, Zhonglin Chen, Yong Ji, Luo |
| author_facet | Xi, Ningyuan Wang, Xiaoyu Wu, Yetao Chen, Teng Gu, Qingqing Zhao, Yue Qu, Jinxian Jiang, Zhonglin Chen, Yong Ji, Luo |
| contents | Current research efforts are focused on enhancing the thinking and reasoning capability of large language model (LLM) by prompting, data-driven emergence and inference-time computation. In this study, we consider stimulating language model's thinking and cognitive abilities from a modular perspective, which mimics the human brain architecture. We select a specific intermediate attention layer with newly implemented language heads. We conduct dual-layer fine-tuning by annotated (query, thought, answer) samples and show that the intermediate layer can also learn to decode fluent and reasonable language tokens. A two-pass inference mechanism is designed to generate thoughts then formal responses. The entire framework is called modularized thinking language model (MeTHanol) which can enhance LLM's cognitive behaviors as indicated by Theory of Mind (ToM) and Vignette-based experiments. Case studies also show that MeTHanol can plan and self-reflect and generate human-like thoughts and answers, even on unseen and open-domain tasks. MeTHanol can also adapt to a personalized prompt and behave as the specified character. Our study holds promise for significant cognitive gains from a modular perspective. Our code, model and data are available at https://bachozean.github.io/methanol-page |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12059 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MeTHanol: Modularized Thinking Language Models with Intermediate Layer Thinking, Decoding and Bootstrapping Reasoning Xi, Ningyuan Wang, Xiaoyu Wu, Yetao Chen, Teng Gu, Qingqing Zhao, Yue Qu, Jinxian Jiang, Zhonglin Chen, Yong Ji, Luo Computation and Language Artificial Intelligence Machine Learning Current research efforts are focused on enhancing the thinking and reasoning capability of large language model (LLM) by prompting, data-driven emergence and inference-time computation. In this study, we consider stimulating language model's thinking and cognitive abilities from a modular perspective, which mimics the human brain architecture. We select a specific intermediate attention layer with newly implemented language heads. We conduct dual-layer fine-tuning by annotated (query, thought, answer) samples and show that the intermediate layer can also learn to decode fluent and reasonable language tokens. A two-pass inference mechanism is designed to generate thoughts then formal responses. The entire framework is called modularized thinking language model (MeTHanol) which can enhance LLM's cognitive behaviors as indicated by Theory of Mind (ToM) and Vignette-based experiments. Case studies also show that MeTHanol can plan and self-reflect and generate human-like thoughts and answers, even on unseen and open-domain tasks. MeTHanol can also adapt to a personalized prompt and behave as the specified character. Our study holds promise for significant cognitive gains from a modular perspective. Our code, model and data are available at https://bachozean.github.io/methanol-page |
| title | MeTHanol: Modularized Thinking Language Models with Intermediate Layer Thinking, Decoding and Bootstrapping Reasoning |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2409.12059 |