Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Beyond Post-Hoc Temperature Scaling: Bilevel Optimization for LLM Calibration
Ruochen Jin, Zhanliang Wang, Zongyu Dai +2
Preference alignment often makes large language models (LLMs) overconfident and poorly calibrated. Traditional post-hoc temperature scaling is inherently domain-dependent: a temper…
cs.LG2025
Restoring Calibration for Aligned Large Language Models: A Calibration-Aware Fine-Tuning Approach
Jiancong Xiao, Bojian Hou, Zhanliang Wang +4
One of the key technologies for the success of Large Language Models (LLMs) is preference alignment. However, a notable side effect of preference alignment is poor calibration: whi…
cs.LG2024
Fine-Tuning Attention Modules Only: Enhancing Weight Disentanglement in Task Arithmetic
Ruochen Jin, Bojian Hou, Jiancong Xiao +2
In recent years, task arithmetic has garnered increasing attention. This approach edits pre-trained models directly in weight space by combining the fine-tuned weights of various t…