PaperHub

Chao Du

~Chao_Du1

39
论文总数
19.5
年均投稿
5.9
平均评分
接收情况26/39
会议分布
ICLR
22
NeurIPS
9
ICML
5
COLM
3

发表论文 (39 篇)

202526

5.8
4

Improving Long-Text Alignment for Text-to-Image Diffusion Models

ICLR 2025Poster
5.0
4

Sample Efficient Alignment for LLMs

ICLR 2025Rejected
6.0
4

A Closer Look at Machine Unlearning for Large Language Models

ICLR 2025Poster
4.5
4

Test-Time Backdoor Attacks on Multimodal Large Language Models

ICLR 2025withdrawn
7.3
3

When Attention Sink Emerges in Language Models: An Empirical View

ICLR 2025Spotlight
4.0
3

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts

ICLR 2025withdrawn
4.0
4

Denial-of-Service Poisoning Attacks against Large Language Models

ICLR 2025withdrawn
7.2
4

Improving Your Model Ranking on Chatbot Arena by Vote Rigging

ICML 2025Poster
6.0
4

Bootstrapping Language Models with DPO Implicit Rewards

ICLR 2025Poster
7.2
4

Continual Reinforcement Learning by Planning with Online World Models

ICML 2025Spotlight
6.5
4

Scaling up Masked Diffusion Models on Text

ICLR 2025Poster
5.3
4

SimLayerKV: A Simple Framework for Layer-Level KV Cache Reduction

ICLR 2025Rejected
7.8
8

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates

ICLR 2025Oral
6.3
4

Improved Techniques for Optimization-Based Jailbreaking on Large Language Models

ICLR 2025Poster
5.8
3

Sample-Efficient Alignment for LLMs

NeurIPS 2025Rejected
6.4
4

Lifelong Safety Alignment for Language Models

NeurIPS 2025Poster
7.3
4

NoisyRollout: Reinforcing Visual Reasoning with Data Augmentation

NeurIPS 2025Poster
6.4
4

Optimizing Anytime Reasoning via Budget Relative Policy Optimization

NeurIPS 2025Poster
5.5
4

Orient Anything: Learning Robust Object Orientation Estimation from Rendering 3D Models

ICML 2025Poster
5.4
5

Weak-to-Strong Jailbreaking on Large Language Models

ICLR 2025Rejected
6.3
3

Weak-to-Strong Jailbreaking on Large Language Models

ICML 2025Poster
6.0
4

LLM-based Multi-Agents System Attack via Continuous Optimization with Discrete Efficient Search

COLM 2025Poster
6.8
4

Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment

NeurIPS 2025Poster
8.7
4

Orient Anything V2: Unifying Orientation and Rotation Understanding

NeurIPS 2025Spotlight
8.0
3

Understanding R1-Zero-Like Training: A Critical Perspective

COLM 2025Poster
4.8
3

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms

ICML 2025Poster

202413