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
69.23/100
前 2.8%
全站排名 #1,789
发表论文15 篇
平均评分
年均产出5.0 篇/年
Xun Wu
研究方向
language-based multi-modal intelligence : multi-modal learning (acoustic · vision · language and point clouds modalities · etc.) · multi-modal retrieval and summarization · downstream adaption (zero-shot learning · parameter-efficient fine-tuning) · unified vision model for diverse tasks (object detection · segmentation and reconstruction
28
VisCodex: Unified Multimodal Code Generation via Merging Vision and Coding Models
ICLR 2026Poster
三作15
Geometric-Mean Policy Optimization
ICLR 2026Poster
27
Code Aesthetics with Agentic Reward Feedback
ICLR 2026Poster
10
BitNet Distillation
ICLR 2026Rejected
一作5
Towards Stable and Effective Reinforcement learning for Mixture-of-Experts
ICLR 2026Withdrawn
二作5
On-Policy RL with Optimal Reward Baseline
ICLR 2026Withdrawn
三作23
Think Only When You Need with Large Hybrid-Reasoning Models
NeurIPS 2025Poster
二作29
Reward Reasoning Models
NeurIPS 2025Poster
8
UNComp: Uncertainty-Aware Long-context Compressor for Efficient Large Language Model Inference
ICLR 2025Rejected
30
UncertaintyRAG: Span Uncertainty Enhanced Long-Context Modeling for Retrieval-Augmented Generation
ICLR 2025Rejected
23
Textual Aesthetics in Large Language Models
ICLR 2025Withdrawn
三作