Think Clearly: Improving Reasoning via Redundant Token Pruning

Daewon Choi, Jimin Lee, Jihoon Tack, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Bhavana Ganesh, Jinwoo Shin, Aram Galstyan, Sravan Babu Bodapati


Abstract
Recent large language models have shown promising capabilities in long-form reasoning, following structured chains of thought before arriving at a final answer. However, we observe that these reasoning paths tend to include substantial redundancy; analyzing attention patterns reveals that attention scores are widely scattered, particularly incorrect answers exhibit greater attention sparsity. In this paper, we demonstrate that deliberately removing this redundancy in the reasoning process significantly improves the performance through clear thinking (i.e., removing distraction). Specifically, we systematically identify such redundancy by measuring token-level attention scores to a special end-of-thinking token, which is appended to an explicit instruction inserted to conclude each intermediate reasoning step. Furthermore, we propose structure-aware pruning that prioritizes removing tokens in low-contributing reasoning chunks over individual tokens. After evicting redundant tokens, we remove the injected end-of-thinking instruction, then resume the reasoning generation. We demonstrate that our method significantly improves the over all accuracy across reasoning-intensive benchmarks without any training involved. In particular, our method shows strong performance on challenging mathematics competition benchmarks such as AIME and AMC, where reasoning redundancy is more prevalent.
Anthology ID:
2025.findings-emnlp.1169
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2025
Month:
November
Year:
2025
Address:
Suzhou, China
Editors:
Christos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
21437–21451
Language:
URL:
https://aclanthology.org/2025.findings-emnlp.1169/
DOI:
Bibkey:
Cite (ACL):
Daewon Choi, Jimin Lee, Jihoon Tack, Woomin Song, Saket Dingliwal, Sai Muralidhar Jayanthi, Bhavana Ganesh, Jinwoo Shin, Aram Galstyan, and Sravan Babu Bodapati. 2025. Think Clearly: Improving Reasoning via Redundant Token Pruning. In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 21437–21451, Suzhou, China. Association for Computational Linguistics.
Cite (Informal):
Think Clearly: Improving Reasoning via Redundant Token Pruning (Choi et al., Findings 2025)
Copy Citation:
PDF:
https://aclanthology.org/2025.findings-emnlp.1169.pdf
Checklist:
 2025.findings-emnlp.1169.checklist.pdf