8 citations · 8 across the 6 of their papers we have counts for
6 papers
Search for Truth from Reasoning: A Dynamic Representation Editing Framework for Steering LLM Trajectories
Tianlong Wang, Yuhang Wang, Weibin Liao +5
Current approaches to enhance Large Language Model (LLM) reasoning, such as Chain-of-Thought and "Wait" prompts, primarily encourage models to think more, yet often fail to guide t…
Beyond Imitation: Reflective On-Policy Self-Distillation for LLM Reasoning
Ziqi Zhao, Xinyu Ma, Liu Yang +6
On-policy self-distillation (OPSD) improves the reasoning capabilities of large language models (LLMs) by providing dense token-level supervision for on-policy rollouts. However, e…
DRESSing Up LLM: Efficient Stylized Question-Answering via Style Subspace Editing
Xinyu Ma, Yifeng Xu, Yang Lin +5
We introduce DRESS, a novel approach for generating stylized large language model (LLM) responses through representation editing. Existing methods like prompting and fine-tuning ar…
Enhancing Topic Interpretability for Neural Topic Modeling through Topic-wise Contrastive Learning
Xin Gao, Yang Lin, Ruiqing Li +4
Data mining and knowledge discovery are essential aspects of extracting valuable insights from vast datasets. Neural topic models (NTMs) have emerged as a valuable unsupervised too…
LoRA Dropout as a Sparsity Regularizer for Overfitting Control
Yang Lin, Xinyu Ma, Xu Chu +4
Parameter-efficient fine-tuning methods, represented by LoRA, play an essential role in adapting large-scale pre-trained models to downstream tasks. However, fine-tuning LoRA-serie…
Parameter Efficient Quasi-Orthogonal Fine-Tuning via Givens Rotation
Xinyu Ma, Xu Chu, Zhibang Yang +3
With the increasingly powerful performances and enormous scales of pretrained models, promoting parameter efficiency in fine-tuning has become a crucial need for effective and effi…