7 papers
CommitKV: Lifecycle-Aware KV Cache Compression via Commit Transitions for Multi-Turn Agents
Weizhong Huang, Jinchao Zhang, Xiawu Zheng
Multi-turn Reasoning-and-Acting (ReAct) agents accumulate growing trajectories of reasoning, tool calls, and observations. Their key-value (KV) caches grow accordingly, increasing…
Neural Langevin Machine: a local asymmetric learning rule can be creative
Zhendong Yu, Weizhong Huang, Haiping Huang
Fixed points of recurrent neural networks can be leveraged to store and generate information. These fixed points can be captured by the Boltzmann-Gibbs measure, which leads to neur…
QuoTA: Query-oriented Token Assignment via CoT Query Decouple for Long Video Comprehension
Yongdong Luo, Wang Chen, Xiawu Zheng +8
Recent advances in long video understanding typically mitigate visual redundancy through visual token pruning based on attention distribution. However, while existing methods emplo…
Freezing chaos without synaptic plasticity
Weizhong Huang, Haiping Huang
Chaos is ubiquitous in high-dimensional neural dynamics. A strong chaotic fluctuation may be harmful to information processing. A traditional way to mitigate this issue is to intro…
Dynamic Low-Rank Sparse Adaptation for Large Language Models
Weizhong Huang, Yuxin Zhang, Xiawu Zheng +4
Despite the efficacy of network sparsity in alleviating the deployment strain of Large Language Models (LLMs), it endures significant performance degradation. Applying Low-Rank Ada…
Determining Layer-wise Sparsity for Large Language Models Through a Theoretical Perspective
Weizhong Huang, Yuxin Zhang, Xiawu Zheng +2
In this paper, we address the challenge of determining the layer-wise sparsity rates of large language models (LLMs) through a theoretical perspective. Specifically, we identify a…