Chengquan Jiang — accepted AI-conference papers

2026

Attn-QAT: 4-Bit Attention With Quantization-Aware Training

Peiyuan Zhang, Matthew Noto, Wenxuan Tan, Chengquan Jiang, Will Lin, Wei Zhou +1

ICML 2026poster

Achieving reliable 4-bit attention is a prerequisite for end-to-end FP4 computation on emerging FP4-capable GPUs, yet attention remains the main obstacle due to FP4's tiny dynamic range and attention's heavy-tailed activations. This paper presents the first systematic study of 4-bit quantization-awa…

Cited by 0 Source ↗Scholar ↗

2026

ReTool: Reinforcement Learning for Strategic Tool Use in LLMs

Jiazhan Feng, Shijue Huang, Xingwei Qu, Ge Zhang, Yujia Qin, Baoquan Zhong +3

ICLR 2026poster

While reasoning models trained with reinforcement learning (RL) excel in reasoning, they struggle in scenarios requiring structured problem-solving, such as geometric reasoning, concise computation, or complex equation solving—areas where computational tools like code interpreters (CI) demonstrate d…

Cited by 0 Source ↗code ↗Scholar ↗