Yizhe Xiong (熊翊哲)

Research

I'm interested in the efficient adaptation of foundation models to diverse downstream tasks through transfer learning. My research focuses on developing methods that enable these large-scale models to generalize effectively across various domains, even when faced with limited data or computational resources. I aim to make advanced AI models more accessible and applicable to a wide range of real-world applications. For a full list of my publications, please visit my Google Scholar page.

Selected Conference Articles:

1. Temporal Scaling Law for Large Language Models
   Yizhe Xiong, Xiansheng Chen, Xin Ye, Hui Chen, Zijia Lin, Haoran Lian, Zhenpeng Su, Wei Huang, Jianwei Niu, Jungong Han, Guiguang Ding
   Keywords: Large Language Model (LLM), Language Modeling, Scaling Law
   EMNLP 2025 (Oral Presentation) | paper

2. PYRA: Parallel Yielding Re-Activation for Training-Inference Efficient Task Adaptation
   Yizhe Xiong, Hui Chen, Tianxiang Hao, Zijia Lin, Jungong Han, Yuesong Zhang, Guoxin Wang, Yongjun Bao, Guiguang Ding
   Keywords: Transfer Learning, Parameter-Efficient Fine-Tuning (PEFT), Task Adaptation, Model Pruning, Token Pruning
   ECCV 2024 | paper  GitHub repo

3. Confidence-based Visual Dispersal for Few-shot Unsupervised Domain Adaptation
   Yizhe Xiong, Hui Chen, Zijia Lin, Sicheng Zhao, Guiguang Ding
   Keywords: Domain Adaptation, Transfer Learning, Few-Shot (Low Shot), Few-Shot Unsupervised Domain Adaptation (FUDA)
   ICCV 2023 | paper  GitHub repo

4. Fast Quiet-STaR: Thinking Without Thought Tokens
   Wei Huang*, Yizhe Xiong*, Xin Ye*, Zhijie Deng, Hui Chen, Zijia Lin, Guiguang Ding (∗ denotes equal contribution)
   Keywords: Large Language Model (LLM), Language Modeling, Reasoning, Reinforcement Learning
   EMNLP 2025 Findings | paper

5. Breaking the Stage Barrier: A Novel Single-Stage Approach to Long Context Extension for Large Language Models
   Haoran Lian∗, Junmin Chen∗, Wei Huang∗, Yizhe Xiong∗, Wenping Hu*, Guiguang Ding, Hui Chen, Jianwei Niu, Zijia Lin, Fuzheng Zhang, Di Zhang (∗ denotes equal contribution)
   Keywords: Large Language Model (LLM), Language Modeling, Long Context Extrapolation
   COLING 2025 | paper

6. LBPE: Long-token-first Tokenization to Improve Large Language Models
   Haoran Lian, Yizhe Xiong, Zijia Lin, Jianwei Niu, Shasha Mo, Hui Chen, Peng Liu, Guiguang Ding
   Keywords: Large Language Model (LLM), Language Modeling, Machine Translation, Byte-Pair Encoding (BPE)
   ICASSP 2025 | paper

7. Scaffold-BPE: Enhancing Byte Pair Encoding for Large Language Models with Simple and Effective Scaffold Token Removal
   Haoran Lian, Yizhe Xiong, Jianwei Niu, Shasha Mo, Zhenpeng Su, Zijia Lin, Peng Liu, Hui Chen, Guiguang Ding
   Keywords: Large Language Model (LLM), Language Modeling, Machine Translation, Byte-Pair Encoding (BPE)
   AAAI 2025 | paper

Journal Articles:

1. Cross-Modality Prompts: Few-shot Multi-label Recognition with Single-label Training
   Zixuan Ding, Zihan Zhou, Hui Chen, Tianxiang Hao, Yizhe Xiong, Sicheng Zhao, Qiang Zhang, Jungong Han
   Keywords: Multi-label Recognition, Transfer Learning, Few-Shot (Low Shot)
   IEEE Transactions on Multimedia | paper

Other Selected Works:

1. UniAttn: Reducing Inference Costs via Softmax Unification for Post-Training LLMs
   Yizhe Xiong, Wei Huang, Xin Ye, Hui Chen, Zijia Lin, Haoran Lian, Zhenpeng Su, Jungong Han, Guiguang Ding
   Keywords: Large Language Model (LLM), Language Modeling, Model Compression, Post-Training, Supervised Fine-Tuning (SFT)
   Under review | paper

2. Neutralizing Token Aggregation via Information Augmentation for Efficient Test-Time Adaptation
   Yizhe Xiong, Zihan Zhou, Yiwen Liang, Hui Chen, Zijia Lin, Tianxiang Hao, Fan Zhang, Jungong Han, Guiguang Ding
   Keywords: Test-Time Training, Vision Transformers, Model Compression
   Under review | paper

Awards