Preprint
Jul 2026
Efficient Chain-of-Modality Reasoning via Progressive Compression for Spoken Language Models
This work proposes Efficient Chain-of-Modality Reasoning (ECoM Reasoning), the first framework to introduce compressed reasoning into SLMs, and proposes Progressive Compression, a curriculum-based strategy that gradually trains the model from full-form reasoning to compressed reasoning.
Pengchao Feng, Chao-Hong Tan, Qian Chen et al.
· 0 citations