activity
20242026
collaborators

25 papers

cs.CL2026

Erase-then-Delta Attention: Decoupling Erase and Write Addresses in Delta-Rule Linear Attention

Xiao Li, Chengruidong Zhang, Hao Luo +15

Delta-rule linear attention improves recurrent memory updates by correcting what is already stored at the current write address before writing new content. However, the active corr…

cs.LG2026

Breaking Entropy Bounds: Accelerating RL Training via MTP with Rejection Sampling

Yucheng Li, Huiqiang Jiang, Yang Xu +14

Reinforcement learning (RL) has become a key component in modern large language models, yet the rollout stage remains the key bottleneck in RL training pipelines. Although Multi-To…

cs.LG2026

ParisKV: Fast and Drift-Robust KV-Cache Retrieval for Long-Context LLMs

Yanlin Qi, Xinhang Chen, Huiqiang Jiang +3

KV-cache retrieval is essential for long-context LLM inference, yet existing methods struggle with distribution drift and high latency at scale. We introduce ParisKV, a drift-robus…

cs.CL2026

xKV: Cross-Layer KV-Cache Compression via Aligned Singular Vector Extraction

Chi-Chih Chang, Wei-Cheng Lin, Chien-Yu Lin +8

Long-context Large Language Models (LLMs) enable powerful applications but incur high memory costs due to the key-value states (KV-Cache). Recent studies attempt to share KV-Cache…

cs.CL2026

MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training

Wenxuan Li, Chengruidong Zhang, Huiqiang Jiang +3

The adoption of long context windows has become a standard feature in Large Language Models (LLMs), as extended contexts significantly enhance their capacity for complex reasoning…

cs.LG2026

DISA: Offline Importance Sampling for Distribution-Matching LLM-RL

Shaobo Wang, Yujie Chen, Yafeng Sun +9

Modern reasoning agents are increasingly evaluated on their ability to generate multiple valid solution paths, plans, or tool-use traces for a given input. Standard reward-maximizi…