papers

Publications (7)

cs.LG2026

Spectral Rewiring for Exploration, Purification, and Model Merging

Zhilong Zhang, Hongli Yu, Huan-ang Gao +5

Reinforcement learning has become a standard post-training recipe for large language models, but dense full-parameter updates create two deployment-relevant bottlenecks: suppressed…

cs.CL2025

Seed1.5-Thinking: Advancing Superb Reasoning Models with Reinforcement Learning

ByteDance Seed, :, Jiaze Chen +267

We introduce Seed1.5-Thinking, capable of reasoning through thinking before responding, resulting in improved performance on a wide range of benchmarks. Seed1.5-Thinking achieves 8…

cs.CL2025

MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent

Hongli Yu, Tinghong Chen, Jiangtao Feng +8

Despite improvements by length extrapolation, efficient attention and memory modules, handling infinitely long documents with linear complexity without performance degradation duri…

cs.CL2025

Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference

Yuxuan Song, Zheng Zhang, Cheng Luo +19

We present Seed Diffusion Preview, a large-scale language model based on discrete-state diffusion, offering remarkably fast inference speed. Thanks to non-sequential, parallel gene…

cs.CL2025

Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles

Jiangjie Chen, Qianyu He, Siyu Yuan +9

Large Language Models (LLMs), such as OpenAI's o1 and DeepSeek's R1, excel at advanced reasoning tasks like math and coding via Reinforcement Learning with Verifiable Rewards (RLVR…

cs.LG2025

DAPO: An Open-Source LLM Reinforcement Learning System at Scale

Qiying Yu, Zheng Zhang, Ruofei Zhu +32

Inference scaling empowers LLMs with unprecedented reasoning ability, with reinforcement learning as the core technique to elicit complex reasoning. However, key technical details…

cs.LG2026

CUDA Agent: Large-Scale Agentic RL for High-Performance CUDA Kernel Generation

Weinan Dai, Hanlin Wu, Qiying Yu +13

GPU kernel optimization is fundamental to modern deep learning but remains a highly specialized task requiring deep hardware expertise. Despite strong performance in general progra…