A POI-aware contrastive training framework using LLM-generated near-misses reduces both general and CS-aware error rates by over 2% on cmn-eng and vie-eng code-switching ASR datasets compared to standard LoRA fine-tuning.
mozillazg/python-pinyin: v0.55.0,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Contrastive Training with LLM-generated Near-Misses for Robust Code-Switching Speech Recognition
A POI-aware contrastive training framework using LLM-generated near-misses reduces both general and CS-aware error rates by over 2% on cmn-eng and vie-eng code-switching ASR datasets compared to standard LoRA fine-tuning.