11 papers
Efficient Long-context Language Model Training by Core Attention Disaggregation
Yonghao Zhuang, Junda Chen, Bo Pang +6
We present core attention disaggregation (CAD), a technique that improves long-context large language model training by decoupling the core attention computation, softmax(QK^T)V, f…
K2-Think: A Parameter-Efficient Reasoning System
Zhoujun Cheng, Richard Fan, Shibo Hao +28
K2-Think is a reasoning system that achieves state-of-the-art performance with a 32B parameter model, matching or surpassing much larger models like GPT-OSS 120B and DeepSeek v3.1.…
Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective
Zhoujun Cheng, Shibo Hao, Tianyang Liu +21
Reinforcement learning (RL) has emerged as a promising approach to improve large language model (LLM) reasoning, yet most open efforts focus narrowly on math and code, limiting our…
LLM360 K2: Building a 65B 360-Open-Source Large Language Model from Scratch
Zhengzhong Liu, Bowen Tan, Hongyi Wang +22
We detail the training of the LLM360 K2-65B model, scaling up our 360-degree OPEN SOURCE approach to the largest and most powerful models under project LLM360. While open-source LL…
Efficiently Scaling LLM Reasoning with Certaindex
Yichao Fu, Junda Chen, Siqi Zhu +8
Test-time reasoning algorithms such as chain-of-thought, self-consistency, and MCTS enhance LLM problem-solving but can wastefully generate many tokens without improving accuracy.…
Helix: Serving Large Language Models over Heterogeneous GPUs and Network via Max-Flow
Yixuan Mei, Yonghao Zhuang, Xupeng Miao +3
This paper introduces Helix, a distributed system for high-throughput, low-latency large language model (LLM) serving in heterogeneous GPU clusters. The key idea behind Helix is to…