activity
20232026
most citedDeepSpeed-Chat: Easy, Fast and Affordable RLHF Training of ChatGPT-like Models at All Scales

10 citations · 32 across the 22 of their papers we have counts for

collaborators

22 papers

cs.LG2026

Tail-Likelihood Reinforcement Learning

Shrinivas Ramasubramanian, Daman Arora, Fahim Tajwar +11

Reinforcement learning typically optimizes average reward. For generative policies, the average can hide an important distinction: two policies can achieve the same mean reward whi…

cs.CL2026

Osprey: Target-agnostic Pre-training Makes Stronger Drafters in Speculative Decoding

Fengxiang Bie, Yuqing Jian, Yifan Yu +7

Speculative decoding is critical for accelerating LLM inference. However, the speedup is fragile: drafters are typically trained against a narrow distribution for a single target m…

cs.CL2026

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

Zunhai Su, Bohan Sun, Xialie Zhuang +8

We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediate…

cs.LG2026

Linear RNN Scaling Laws: When Longer Sequences Beat More Sequences

Ziyan Chen, Zhongzhu Zhou, Peilin Liu +1

Empirical scaling laws for autoregressive language models relate prediction loss to model size, data size, and optimization compute, but their theoretical origin is still poorly un…

cs.LG2026

Sketched Linear Contrastive Learning: Approximation, Optimization, and Statistical Scaling

Ziyan Chen, Zhongzhu Zhou, Ding-Xuan Zhou

Scaling laws describe how learning performance varies with model size, data size, and compute. While recent theoretical work has established scaling laws for sketched linear regres…

cs.LG2026

Taylor-Calibrate: Principled Initialization for Hybrid Linear Attention Distillation

Zhongzhu Zhou, Qingyang Wu, Junxiong Wang +4

Hybrid linear attention models offer an appealing path to faster long-context inference: they reduce the quadratic cost and KV-cache burden of full softmax attention while retainin…