影响力指数

50.57/100

前 8.3%

全站排名 #5,356

发表论文15 篇

平均评分5.4

年均产出5.0 篇/年

Lianli Gao

Full Professor@University of Electronic Science and Technology of China·中国·OpenReview

研究方向

AI safty · multimedia analysis · Image analysis · Video analysis · Vison and Language · Adversarial Attack · Few-shot Learning · Scene graph generation · Pose estimation · visual synthesis · Visual Retrieval · agentic AI · Visual Generation

Policy Contrastive Decoding for Robotic Foundation Models

ICLR 2026Poster

Prototype-based Regularization Learning For Text-Video Retrieval

ICLR 2026Rejected

Dynamic Pattern Alignment Learning for Pretraining Lightweight Human-Centric Vision Models

ICLR 2026Withdrawn

Stable-FAST: Stabilizing Inference of Autoregressive Vision-Language-Action Models

ICLR 2026Withdrawn

SafePTR: Token-Level Jailbreak Defense in Multimodal LLMs via Prune-then-Restore Mechanism

NeurIPS 2025Poster

Safe + Safe = Unsafe? Exploring How Safe Images Can Be Exploited to Jailbreak Large Vision-Language Models

NeurIPS 2025Poster

FlexAC: Towards Flexible Control of Associative Reasoning in Multimodal Large Language Models

NeurIPS 2025Poster

Debiased Orthogonal Boundary-driven Efficient Noise Mitigation

ICML 2025Rejected

MMEvol: Empowering Multimodal Large Language Models with Evol-Instruct

ICLR 2025Rejected

SeMv-3D: Towards Semantic and Mutil-view Consistency simultaneously for General Text-to-3D Generation with Triplane Priors

ICLR 2025Rejected

One-step Noisy Label Mitigation

ICLR 2025Withdrawn

ETC: Towards Training-Efficient Video Synthesis with Exploiting Temporal Capabilities of Spatial Attention

ICLR 2025Withdrawn

合作者 (20)