5 papers
Kimi K3: Open Frontier Intelligence
Kimi Team, Tongtong Bai, Yifan Bai +398
We introduce Kimi K3, a 2.8T parameter Mixture-of-Experts model with 104 billion activated parameters, native vision capabilities, and a 1-million-token context window. Kimi K3 is…
EntroRouter: Learning Efficient Model Routing via Entropy Regulation
Kaiyi Zhang, Xueliang Zhao, Zhuocheng Gong +2
Model routing balances solution accuracy and computational cost by selecting among models of varying capabilities. While recent multi-round frameworks interleave reasoning and plan…
DelTA: Discriminative Token Credit Assignment for Reinforcement Learning from Verifiable Rewards
Kaiyi Zhang, Wei Wu, Yankai Lin
Reinforcement learning from verifiable rewards (RLVR) has emerged as a central technique for improving the reasoning capabilities of large language models. Despite its effectivenes…
How GenAI Mentor Configurations Shape Early Collaborative Dynamics: A Classroom Comparison of Individual and Shared Agents
Siyu Zha, Weijing Liu, Fei Qin +6
Generative artificial intelligence (GenAI) is increasingly embedded in computer-supported collaborative learning (CSCL), yet little empirical research has unpacked how different co…
StepHint: Multi-level Stepwise Hints Enhance Reinforcement Learning to Reason
Kaiyi Zhang, Ang Lv, Jinpeng Li +4
Reinforcement learning with verifiable rewards (RLVR) is a promising approach for improving the complex reasoning abilities of large language models (LLMs). However, current RLVR m…