1 citations · 2 across the 9 of their papers we have counts for
11 papers
Selective Channel Restoration for Backdoored Vision-Language Models
Shuming Liu, Zhifang Zhang, Suqin Yuan +3
Vision-language models (VLMs) exhibit strong multimodal capabilities but remain vulnerable to backdoors implanted through poisoned fine-tuning data. Existing defenses often require…
Decoupling Token Roles in Autoregressive Pretraining
Suqin Yuan, Runqi Lin, Kevin Qinghong Lin +4
Autoregressive pretraining increasingly draws on heterogeneous data, making it important to understand how a model learns from an individual token. The next-token prediction object…
Are Human-Aligned Models Models of Humans? A Turing-Test Gap in Preference Alignment
Suqin Yuan, Runqi Lin, Muyang Li +5
Human-feedback alignment has made language models useful assistants and is commonly described as aligning them with humans. However, the responses people prefer from an AI need not…
PaperRouter-Agent: A Content-Grounded LLM Agent for Personalized Hierarchical Paper Routing
Keshen Zhou, Lintao Wang, Suqin Yuan +3
Researchers organize the papers they collect into personal folder hierarchies in reference managers, and route each new paper into the folder where it belongs. This task differs fr…
Understanding Diversity Collapse in RLVR via the Lens of Overtraining
Suqin Yuan, Jinkun Chen, Jiyang Zheng +6
Reinforcement learning with verifiable rewards (RLVR) has become a key approach for enhancing the reasoning abilities of large language models. However, RLVR often suffers from \em…
Mitigating Mismatch within Reference-based Preference Optimization
Suqin Yuan, Xingrui Yu, Jiyang Zheng +4
Direct Preference Optimization (DPO) has become the de facto standard for offline preference alignment of large language models, but its reliance on a reference policy introduces a…
FORCE: Transferable Visual Jailbreaking Attacks via Feature Over-Reliance CorrEction
Runqi Lin, Alasdair Paren, Suqin Yuan +4
The integration of new modalities enhances the capabilities of multimodal large language models (MLLMs) but also introduces additional vulnerabilities. In particular, simple visual…
Early Stopping Against Label Noise Without Validation Data
Suqin Yuan, Lei Feng, Tongliang Liu
Early stopping methods in deep learning face the challenge of balancing the volume of training and validation data, especially in the presence of label noise. Concretely, sparing m…
Instance-dependent Early Stopping
Suqin Yuan, Runqi Lin, Lei Feng +2
In machine learning practice, early stopping has been widely used to regularize models and can save computational costs by halting the training process when the model's performance…
Enhancing Sample Selection Against Label Noise by Cutting Mislabeled Easy Examples
Suqin Yuan, Lei Feng, Bo Han +1
Sample selection is a prevalent approach in learning with noisy labels, aiming to identify confident samples for training. Although existing sample selection methods have achieved…
Late Stopping: Avoiding Confidently Learning from Mislabeled Examples
Suqin Yuan, Lei Feng, Tongliang Liu
Sample selection is a prevalent method in learning with noisy labels, where small-loss data are typically considered as correctly labeled data. However, this method may not effecti…