作者位次 #2/5ACL 2026Poster· Natural Language Processing
Cut Your Losses! Learning to Prune Paths Early for Efficient Parallel Reasoning
Jiaxi Bi · Tongxu Luo · Wenyu Du · Zhengyang Tang · Benyou Wang
摘要
Parallel reasoning enhances Large Reasoning Models (LRMs) but incurs prohibitive costs due to futile paths caused by early errors. To mitigate this, path pruning at the prefix level is essential, yet existing research remains fragmented without a standardized framework. In this work, we propose t...