影响力指数
论文质量、代表作、近期表现、广度与样本量置信度综合计算
88.78/100
前 0.7%
全站排名 #428
发表论文23 篇
平均评分
年均产出7.7 篇/年
qiang liu
研究方向
machine learning · reinforcement learning · off-policy evaluation · deep learning · approximate inference · variational inference · MCMC · probabilistic inference · bayesian inference · langevin dynamics · Monte Carlo · importance sampling · Stein variational · graphical models · Stein's method
18
Cautious Weight Decay
ICLR 2026Poster
通讯17
Cautious Optimizers: Improving Training with One Line of Code
ICLR 2026Poster
通讯15
$\nabla$-Reasoner: LLM Reasoning via Test-Time Gradient Descent in Latent Space
ICLR 2026Poster
19
Gumbel Distillation for Parallel Text Generation
ICLR 2026Poster
通讯16
The Curious Case of AdamW
ICLR 2026Rejected
三作24
DeMo: Decoupled Momentum Optimization
ICLR 2026Poster
通讯15
Longhorn: State Space Models are Amortized Online Learners
ICLR 2025Poster
通讯8
SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization
ICML 2025Poster
三作11
PIPA: Preference Alignment as Prior-Informed Statistical Estimation
ICML 2025Poster
三作15
Improving Adaptive Moment Optimization via Preconditioner Diagonalization
ICLR 2025Rejected
通讯