APA (7th ed.) Citation

Li, X., Wu, S., & Shen, Z. (2026). Decomposing MXFP4 quantization error for LLM reinforcement learning: Reducible bias, recoverable deadzone, and an irreducible floor.

Chicago Style (17th ed.) Citation

Li, Xiaocan, Shiliang Wu, and Zheng Shen. Decomposing MXFP4 Quantization Error for LLM Reinforcement Learning: Reducible Bias, Recoverable Deadzone, and an Irreducible Floor. 2026.

MLA (9th ed.) Citation

Li, Xiaocan, et al. Decomposing MXFP4 Quantization Error for LLM Reinforcement Learning: Reducible Bias, Recoverable Deadzone, and an Irreducible Floor. 2026.