首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Changhai Zhou · Shiyang Zhang · Yuhua Zhou · Qian Qiao · Jun Gao · Cheng Jin · KAIZHOU QIN · Weizhong Zhang
Lingyin Zhang · Jun Gao · Xiaoxue Ren · Ziqiang Cao
Process Reward Models (PRMs), which assign fine-grained scores to intermediate reasoning steps within a solution trajectory, have emerged as a promising approach to enhance the reasoning quality of Large Language Models (LLMs).However, most existing PRMs rely on a unidirectional left-to-right (L2...
Jun Gao · Yun Peng · Qian Qiao · Changhai Zhou · Yuhua Zhou · Shiyang Zhang · Shichao Weng · Zhenchang Xing · Xiaoxue Ren
Despite strong performance on code generation tasks, it remains unclear whether large language models (LLMs) genuinely reason about code execution. Existing code reasoning benchmarks primarily evaluate final output correctness under a single canonical implementation, leaving two critical aspects...
Changhai Zhou · Shiyang Zhang · Yuhua Zhou · Jun Gao · Qian Qiao · Shichao Weng · WEIZHONG ZHANG · Cheng Jin
Deploying and fine-tuning Large Language Models (LLMs) on resource-constrained edge devices requires navigating a strict trade-off between memory footprint and task performance. Existing quantization-aware fine-tuning methods typically decouple weight precision and adapter capacity, overlooking t...
Yuhua Zhou · Shichao Weng · Changhai Zhou · Yuhan Wu · Qian Qiao · Jun Gao · Fei Yang · Aimin Pan
While the massive scale of modern LLMs enables remarkable performance, their static, input-agnostic computational graph incurs substantial resource wastage and high latency during inference. Existing dynamic schemes, such as early-exit and layer-drop reduce FLOPs but break batch processing or int...