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
80.7/100
前 1.3%
全站排名 #850
发表论文20 篇
平均评分
年均产出6.7 篇/年
Yiming Li
研究方向
Backdoor Attack · Backdoor Defense · AI Copyright Protection · Data Copyright · Model Copyright · Responsible AI · Trustworthy ML
26
AudioTrust: Benchmarking The Multifaceted Trustworthiness of Audio Large Language Models
ICLR 2026Poster
16
Explainable Token-level Noise Filtering for LLM Fine-tuning Datasets
ICLR 2026Poster
5
Imperceptible Jailbreaking against Large Language Models
ICLR 2026Withdrawn
二作5
BitHydra: Towards Bit-flip Inference Cost Attack against Large Language Models
ICLR 2026Withdrawn
二作28
Towards Resilient Safety-driven Unlearning for Diffusion Models against Downstream Fine-tuning
NeurIPS 2025Poster
25
Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
NeurIPS 2025Poster
26
Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment
NeurIPS 2025Poster
29
Backdoor Cleaning without External Guidance in MLLM Fine-tuning
NeurIPS 2025Poster
30
VideoShield: Regulating Diffusion-based Video Generation Models via Watermarking
ICLR 2025Poster
三作33
Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models
ICLR 2025Poster
通讯31
A Benchmark for Semantic Sensitive Information in LLMs Outputs
ICLR 2025Poster
49
REFINE: Inversion-Free Backdoor Defense via Model Reprogramming
ICLR 2025Poster
13
Cowpox: Towards the Immunity of VLM-based Multi-Agent Systems
ICML 2025Poster
三作36
Towards False-claim-resistant Model Ownership Verification via Targeted Fingerprint
ICLR 2025Rejected
10
SuperMark: Robust and Training-free Image Watermarking via Diffusion-based Super-Resolution
ICLR 2025Rejected
三作5
RAG$^C$: Towards Copyright Protection for Knowledge Bases of Retrieval-augmented Language Models
ICLR 2025Withdrawn
二作