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
24.96/100
前 31.3%
全站排名 #20,161
发表论文10 篇
平均评分
年均产出5.0 篇/年
Shenghua Liu
研究方向
Post training and reinforcement learning · Large Language Model · In context learning · Trustworthy foundation model · Knowedge editing · Big graph mining · Graph summarization
20
RefineX: Learning to Refine Pre-training Data at Scale from Expert-Guided Programs
ICLR 2026Desk Rejected
二作36
Parameters vs. Context: Fine-Grained Control of Knowledge Reliance in Language Models
ICLR 2026Poster
二作12
Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning
ICLR 2026Withdrawn
二作5
a1: Steep Test-time Scaling Law via Environment Augmented Generation
ICLR 2026Withdrawn
二作6
Lost in Time: Systematic Temporal Bias in Large Audio Language Models
ICLR 2026Rejected
二作5
Innate Reasoning is Not Enough: In-Context Learning Enhances Reasoning Large Language Models with Less Overthinking
ICLR 2026Withdrawn
二作5
Probing Latent Knowledge Conflict for Faithful Retrieval-Augmented Generation
ICLR 2026Withdrawn
18
Is Factuality Enhancement a Free Lunch For LLMs? Better Factuality Can Lead to Worse Context-Faithfulness
ICLR 2025Poster
二作27
HiddenGuard: Fine-Grained Safe Generation with Specialized Representation Router
ICLR 2025Rejected
二作6
Decoding by Contrasting Knowledge: Enhancing LLMs' Confidence on Edited Facts
ICLR 2025Withdrawn
二作