T. Ma

collaborators

4 papers

cs.LG2026

BitsMoE: Efficient Spectral Energy-Guided Bit Allocation for MoE LLM Quantization

Jiayu Zhao, Zihan Teng, Minhao Fan +4

Mixture-of-Experts (MoE) large language models reduce per-token computation through sparse expert activation, but their deployment remains memory-intensive because all expert weigh…

hep-ex2026

Observation of the $X (2370)$ in $J / ψ \to γ K_{S}^{0} K_{S}^{0} π^{0}$ and $J / ψ \to γ π^{0} π^{0} η$

BESIII Collaboration, M. Ablikim, M. N. Achasov +769

Based on $(10087 \pm 44) \times 1 0^{6}$ $J / ψ$ events collected with the BESIII detector, the $J / ψ \to γ K_{S}^{0} K_{S}^{0} π^{0}$ and $J / ψ \to γ π^{0} π^{0} η$ processes are stu…

cs.AR2025

Hardwired-Neurons Language Processing Units as General-Purpose Cognitive Substrates

Yang Liu, Yi Chen, Yongwei Zhao +24

The rapid advancement of Large Language Models (LLMs) has established language as a core general-purpose cognitive substrate, driving the demand for specialized Language Processing…

cs.AR2025

QiMeng: Fully Automated Hardware and Software Design for Processor Chip

Rui Zhang, Yuanbo Wen, Shuyao Cheng +17

Processor chip design technology serves as a key frontier driving breakthroughs in computer science and related fields. With the rapid advancement of information technology, conven…

BitsMoE: Efficient Spectral Energy-Guided Bit Allocation for MoE LLM Quantization

Observation of the X(2370) in J/ψ→γKS0​KS0​π0 and J/ψ→γπ0π0η

Hardwired-Neurons Language Processing Units as General-Purpose Cognitive Substrates

QiMeng: Fully Automated Hardware and Software Design for Processor Chip

Observation of the $X (2370)$ in $J / ψ \to γ K_{S}^{0} K_{S}^{0} π^{0}$ and $J / ψ \to γ π^{0} π^{0} η$