5 papers
DeepSeek-V4.1-Flash: Pushing the Limits of KV Cache Compression
DeepSeek-AI, :, Anyi Xu +585
The widespread adoption of long-horizon agents has made model workloads increasingly input-heavy. Although prior work has substantially reduced the cost of long-context computation…
Understanding Transformer from the Perspective of Associative Memory
Shu Zhong, Mingyu Xu, Tenglong Ao +1
In this paper, we share our reflections and insights on understanding Transformer architectures through the lens of associative memory--a classic psychological concept inspired by…
Amplify Adjacent Token Differences: Enhancing Long Chain-of-Thought Reasoning with Shift-FFN
Yao Xu, Mingyu Xu, Fangyu Lei +7
Recently, models such as OpenAI-o1 and DeepSeek-R1 have demonstrated remarkable performance on complex reasoning tasks through Long Chain-of-Thought (Long-CoT) reasoning. Although…
LongReD: Mitigating Short-Text Degradation of Long-Context Large Language Models via Restoration Distillation
Zican Dong, Junyi Li, Jinhao Jiang +4
Large language models (LLMs) have gained extended context windows through scaling positional encodings and lightweight continual pre-training. However, this often leads to degraded…
KV Shifting Attention Enhances Language Modeling
Mingyu Xu, Wei Cheng, Bingning Wang +1
The current large language models are mainly based on decode-only structure transformers, which have great in-context learning (ICL) capabilities. It is generally believed that the…