3 papers
cs.CL2026
Lngram: N-gram Conditional Memory in Latent Space
Yunao Zheng, Guoyang Xia, Xiaojie Wang +1
Sequence modeling requires both compositional reasoning and local static knowledge retrieval, yet standard Transformers handle both through dense computation. Engram partially deco…
cs.CL2026
ROSA-Tuning: Enhancing Long-Context Modeling via Suffix Matching
Yunao Zheng, Xiaojie Wang, Lei Ren +1
Long-context capability and computational efficiency are among the central challenges facing today's large language models. Existing efficient attention methods reduce computationa…
cs.CL2025
Enhancing Complex Instruction Following for Large Language Models with Mixture-of-Contexts Fine-tuning
Yuheng Lu, ZiMeng Bai, Caixia Yuan +2
Large language models (LLMs) exhibit remarkable capabilities in handling natural language tasks; however, they may struggle to consistently follow complex instructions including th…