From Chaos to Order: The Atomic Reasoner Framework for Fine-grained Reasoning in Large Language Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916658014060544 |
|---|---|
| author | Liu, Jinyi Zheng, Yan Cheng, Rong Wu, Qiyu Guo, Wei Ni, Fei Liang, Hebin Yuan, Yifu Mao, Hangyu Zhang, Fuzheng Hao, Jianye |
| author_facet | Liu, Jinyi Zheng, Yan Cheng, Rong Wu, Qiyu Guo, Wei Ni, Fei Liang, Hebin Yuan, Yifu Mao, Hangyu Zhang, Fuzheng Hao, Jianye |
| contents | Recent advances in large language models (LLMs) have shown remarkable progress, yet their capacity for logical ``slow-thinking'' reasoning persists as a critical research frontier. Current inference scaling paradigms suffer from two fundamental constraints: fragmented thought flows compromising logical coherence, and intensively computational complexity that escalates with search space dimensions. To overcome these limitations, we present \textbf{Atomic Reasoner} (\textbf{AR}), a cognitive inference strategy that enables fine-grained reasoning through systematic atomic-level operations. AR decomposes the reasoning process into atomic cognitive units, employing a cognitive routing mechanism to dynamically construct reasoning representations and orchestrate inference pathways. This systematic methodology implements stepwise, structured cognition, which ensures logical coherence while significantly reducing cognitive load, effectively simulating the cognitive patterns observed in human deep thinking processes. Extensive experimental results demonstrate AR's superior reasoning capabilities without the computational burden of exhaustive solution searches, particularly excelling in linguistic logic puzzles. These findings substantiate AR's effectiveness in enhancing LLMs' capacity for robust, long-sequence logical reasoning and deliberation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_15944 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | From Chaos to Order: The Atomic Reasoner Framework for Fine-grained Reasoning in Large Language Models Liu, Jinyi Zheng, Yan Cheng, Rong Wu, Qiyu Guo, Wei Ni, Fei Liang, Hebin Yuan, Yifu Mao, Hangyu Zhang, Fuzheng Hao, Jianye Computation and Language Recent advances in large language models (LLMs) have shown remarkable progress, yet their capacity for logical ``slow-thinking'' reasoning persists as a critical research frontier. Current inference scaling paradigms suffer from two fundamental constraints: fragmented thought flows compromising logical coherence, and intensively computational complexity that escalates with search space dimensions. To overcome these limitations, we present \textbf{Atomic Reasoner} (\textbf{AR}), a cognitive inference strategy that enables fine-grained reasoning through systematic atomic-level operations. AR decomposes the reasoning process into atomic cognitive units, employing a cognitive routing mechanism to dynamically construct reasoning representations and orchestrate inference pathways. This systematic methodology implements stepwise, structured cognition, which ensures logical coherence while significantly reducing cognitive load, effectively simulating the cognitive patterns observed in human deep thinking processes. Extensive experimental results demonstrate AR's superior reasoning capabilities without the computational burden of exhaustive solution searches, particularly excelling in linguistic logic puzzles. These findings substantiate AR's effectiveness in enhancing LLMs' capacity for robust, long-sequence logical reasoning and deliberation. |
| title | From Chaos to Order: The Atomic Reasoner Framework for Fine-grained Reasoning in Large Language Models |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.15944 |