Showing cs.CLShow all
2 papers · 1 filter
cs.CL2026
Rethinking Length-Based Training: Batch Composition and Loss Normalization in Speech Token Language Models
Hongjin Song, Runwu Shi, Weiqiao Shan +4
Short-to-long training is a simple curriculum for speech models, but its gains can be difficult to interpret. In speech token language models, length-based training can change the…
cs.CL2026
Locality Matters for Training-Free Audio Token Compression in Audio-Language Models
Jiale Luo, Xiaoyu Liang, Haoji Hu
Audio-language models (ALMs) are increasingly used for audio captioning, question answering, and open-ended audio understanding, but their inference cost remains high when audio in…