activity
20242026
collaborators

5 papers

cs.CL2026

Know When to Stop: Segment-Level Credit Assignment for Reducing Overthinking

Chia-Hsuan Lee, Sihui Dai, Mingyang Zhou +5

Reasoning language models frequently overthink: generating extended chains of behaviors such as hedging, approach abandonment, and self contradiction that consume tokens without im…

eess.AS2026

Pseudo2Real: Task Arithmetic for Pseudo-Label Correction in Automatic Speech Recognition

Yi-Cheng Lin, Yu-Hsuan Li Liang, Hsuan Su +4

Robust ASR under domain shift is crucial because real-world systems encounter unseen accents and domains with limited labeled data. Although pseudo-labeling offers a practical work…

cs.CL2025

Learning to Reason for Hallucination Span Detection

Hsuan Su, Ting-Yao Hu, Hema Swetha Koppula +7

Large language models (LLMs) often generate hallucinations -- unsupported content that undermines reliability. While most prior works frame hallucination detection as a binary task…

cs.CL2025

Jailbreaking with Universal Multi-Prompts

Yu-Ling Hsu, Hsuan Su, Shang-Tse Chen

Large language models (LLMs) have seen rapid development in recent years, revolutionizing various applications and significantly enhancing convenience and productivity. However, al…

eess.AS2024

Task Arithmetic can Mitigate Synthetic-to-Real Gap in Automatic Speech Recognition

Hsuan Su, Hua Farn, Fan-Yun Sun +2

Synthetic data is widely used in speech recognition due to the availability of text-to-speech models, which facilitate adapting models to previously unseen text domains. However, e…