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
71.43/100
前 2.4%
全站排名 #1,537
发表论文18 篇
平均评分
年均产出6.0 篇/年
Zechun Liu
研究方向
Knowledge distillation · Pruning · Neural architecture search · Quantization · Neural network compression
17
MobileLLM-R1: Exploring the Limits of Sub-Billion Language Model Reasoners with Open Training Recipes
ICLR 2026Poster
三作20
STEM: SCALING TRANSFORMERS WITH EMBEDDING MODULES
ICLR 2026Poster
22
WiniQ: Accelerating Quantization-Aware Training of LLMs around Saddle Points
ICLR 2026Rejected
二作19
Why RL Updates Look Sparse: An Implicit Compass Drives Optimization Bias
ICLR 2026Rejected
15
ParetoQ: Improving Scaling Laws in Extremely Low-bit LLM Quantization
NeurIPS 2025Poster
一作11
Agent-as-a-Judge: Evaluate Agents with Agents
ICML 2025Poster
13
LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
ICML 2025Poster
14
Param$\Delta$ for Direct Mixing: Post-Train Large Language Model At Zero Cost
ICLR 2025Poster
通讯21
RDD: Retrieval-Based Demonstration Decomposer for Planner Alignment in Long-Horizon Tasks
NeurIPS 2025Poster
三作11
PARQ: Piecewise-Affine Regularized Quantization
ICML 2025Poster
三作36
SpinQuant: LLM Quantization with Learned Rotations
ICLR 2025Poster
一作28
LongVU: Spatiotemporal Adaptive Compression for Long Video-Language Understanding
ICLR 2025Rejected
29
Agent-as-a-Judge: Evaluating Agents with Agents
ICLR 2025Rejected
28
R-Sparse: Rank-Aware Activation Sparsity for Efficient LLM Inference
ICLR 2025Poster
二作