8 papers
RAM-Net: Expressive Linear Attention with Selectively Addressable Memory
Kaicheng Xiao, Haotian Li, Liran Dong +1
While linear attention architectures offer efficient inference, compressing unbounded history into a fixed-size memory inherently limits expressivity and causes information loss. T…
Yunjue Agent Tech Report: A Fully Reproducible, Zero-Start In-Situ Self-Evolving Agent System for Open-Ended Tasks
Haotian Li, Shijun Yang, Weizhen Qi +5
Conventional agent systems often struggle in open-ended environments where task distributions continuously drift and external supervision is scarce. Their reliance on static toolse…
SCULPT: Constraint-Guided Pruned MCTS that Carves Efficient Paths for Mathematical Reasoning
Qitong Fang, Haotian Li, Xu Wang
Automated agent workflows can enhance the problem-solving ability of large language models (LLMs), but common search strategies rely on stochastic exploration and often traverse im…
TrajSelector: Harnessing Latent Representations for Efficient and Effective Best-of-N in Large Reasoning Model
Bin Yu, Xinming Wang, Shijie Lian +6
Large language models (LLMs) have shown remarkable progress in complex reasoning tasks, largely enabled by test-time scaling (TTS) paradigms that allocate additional compute during…
Not All Tokens Are What You Need In Thinking
Hang Yuan, Bin Yu, Haotian Li +6
Modern reasoning models, such as OpenAI's o1 and DeepSeek-R1, exhibit impressive problem-solving capabilities but suffer from critical inefficiencies: high inference latency, exces…
Two Is Better Than One: Rotations Scale LoRAs
Hongcan Guo, Guoshun Nan, Yuan Yang +9
Scaling Low-Rank Adaptation (LoRA)-based Mixture-of-Experts (MoE) facilitates large language models (LLMs) to efficiently adapt to diverse tasks. However, traditional gating mechan…