14 papers
Towards Purified Multi-Label Test-Time Adaptation of Vision-Language Models
Yiwen Liang, Hui Chen, Yizhe Xiong +7
Test-time adaptation (TTA) has been widely explored in single-label recognition, effectively mitigating distribution shifts, especially when combined with vision-language models. H…
Advancing Reliable Test-Time Adaptation of Vision-Language Models under Visual Variations
Yiwen Liang, Hui Chen, Yizhe Xiong +7
Vision-language models (VLMs) exhibit remarkable zero-shot capabilities but struggle with distribution shifts in downstream tasks when labeled data is unavailable, which has motiva…
Neutralizing Token Aggregation via Information Augmentation for Efficient Test-Time Adaptation
Yizhe Xiong, Zihan Zhou, Yiwen Liang +6
Test-Time Adaptation (TTA) has emerged as an effective solution for adapting Vision Transformers (ViT) to distribution shifts without additional training data. However, existing TT…
Fast Quiet-STaR: Thinking Without Thought Tokens
Wei Huang, Yizhe Xiong, Xin Ye +4
Large Language Models (LLMs) have achieved impressive performance across a range of natural language processing tasks. However, recent advances demonstrate that further gains parti…
Finedeep: Mitigating Sparse Activation in Dense LLMs via Multi-Layer Fine-Grained Experts
Leiyu Pan, Zhenpeng Su, Minxuan Lv +10
Large language models have demonstrated exceptional performance across a wide range of tasks. However, dense models usually suffer from sparse activation, where many activation val…
DSMoE: Matrix-Partitioned Experts with Dynamic Routing for Computation-Efficient Dense LLMs
Minxuan Lv, Zhenpeng Su, Leiyu Pan +10
As large language models continue to scale, computational costs and resource consumption have emerged as significant challenges. While existing sparsification methods like pruning…
UniAttn: Reducing Inference Costs via Softmax Unification for Post-Training LLMs
Yizhe Xiong, Wei Huang, Xin Ye +6
Post-training is essential for adapting Large Language Models (LLMs) to real-world applications. Deploying post-trained models faces significant challenges due to substantial memor…
LBPE: Long-token-first Tokenization to Improve Large Language Models
Haoran Lian, Yizhe Xiong, Zijia Lin +5
The prevalent use of Byte Pair Encoding (BPE) in Large Language Models (LLMs) facilitates robust handling of subword units and avoids issues of out-of-vocabulary words. Despite its…
CartesianMoE: Boosting Knowledge Sharing among Experts via Cartesian Product Routing in Mixture-of-Experts
Zhenpeng Su, Xing Wu, Zijia Lin +6
Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the…
MaskMoE: Boosting Token-Level Learning via Routing Mask in Mixture-of-Experts
Zhenpeng Su, Zijia Lin, Xue Bai +8
Scaling the size of a model enhances its capabilities but significantly increases computation complexity. Mixture-of-Experts models (MoE) address the issue by allowing model size t…
Quantized Prompt for Efficient Generalization of Vision-Language Models
Tianxiang Hao, Xiaohan Ding, Juexiao Feng +3
In the past few years, large-scale pre-trained vision-language models like CLIP have achieved tremendous success in various fields. Naturally, how to transfer the rich knowledge in…
More is Better: Deep Domain Adaptation with Multiple Sources
Sicheng Zhao, Hui Chen, Hu Huang +2
In many practical applications, it is often difficult and expensive to obtain large-scale labeled data to train state-of-the-art deep neural networks. Therefore, transferring the l…
Scaffold-BPE: Enhancing Byte Pair Encoding for Large Language Models with Simple and Effective Scaffold Token Removal
Haoran Lian, Yizhe Xiong, Jianwei Niu +7
Byte Pair Encoding (BPE) serves as a foundation method for text tokenization in the Natural Language Processing (NLP) field. Despite its wide adoption, the original BPE algorithm h…
Temporal Scaling Law for Large Language Models
Yizhe Xiong, Xiansheng Chen, Xin Ye +8
Recently, Large Language Models (LLMs) have been widely adopted in a wide range of tasks, leading to increasing attention towards the research on how scaling LLMs affects their per…