3 papers
cs.LG2026
Concept Component Analysis: A Principled Approach for Concept Extraction in LLMs
Yuhang Liu, Erdun Gao, Dong Gong +2
Developing human understandable interpretation of large language models (LLMs) becomes increasingly critical for their deployment in essential domains. Mechanistic interpretability…
cs.AI2025
Causality-aligned Prompt Learning via Diffusion-based Counterfactual Generation
Xinshu Li, Ruoyu Wang, Erdun Gao +2
Prompt learning has garnered attention for its efficiency over traditional model training and fine-tuning. However, existing methods, constrained by inadequate theoretical foundati…
cs.LG2025
On the Value of Cross-Modal Misalignment in Multimodal Representation Learning
Yichao Cai, Yuhang Liu, Erdun Gao +4
Multimodal representation learning, exemplified by multimodal contrastive learning (MMCL) using image-text pairs, aims to learn powerful representations by aligning cues across mod…