1 citations · 3 across the 19 of their papers we have counts for
22 papers
Bridging the Confidence Gap: Temperature Scaling for Calibrating Test-Time Prompt Tuning
Yuwei Liang, Jian Liang, Dapeng Hu +2
Test-time prompt tuning (TPT) enables adaptation on a single test instance, achieving improved accuracy but often sacrificing calibration performance. Most existing calibration met…
To Adapt or Not to Adapt? Selective Adaptation for Vision-Language Models
Siru Jiang, Yuwei Liang, Jian Liang +2
Test-time adaptation (TTA) has emerged as a prominent strategy for adapting vision-language models to distribution shifts during inference. We conduct a per-sample analysis of mode…
AgentStream: How Well Do Self-Evolving LLM Agents Perform Under Streaming Tasks?
Dong Yan, Jian Liang, Dapeng Hu +4
Large language model (LLM) agents can self-evolve by continually improving from their own accumulated experience. However, existing studies predominantly adopt independent evaluati…
USE: A Unified Self-Ensembling Framework for Test-Time Prompt Tuning
Siru Jiang, Jian Liang, Ran He +1
Test-time adaptation (TTA) has emerged as a popular paradigm for improving the performance of vision-language models (e.g., CLIP) on downstream tasks. Among existing CLIP-based TTA…
On the Vulnerability of Parameter-Level Defenses to Model Merging
Kuangpu Guo, Qingyan Zheng, Jian Liang +4
The training-free integration of expert models via model merging has exposed significant security risks, enabling free-riders to combine specialized models without authorization. R…
WorldCoder-Bench: Benchmarking Physically Grounded 3D World Synthesis
Shuo Lu, Yinuo Xu, Kecheng Yu +8
Large language models (LLMs) are increasingly asked not only to write static interfaces, but to construct executable interactive worlds from natural language. Browser-native 3D, co…
Understanding and Mitigating Spurious Signal Amplification in Test-Time Reinforcement Learning for Math Reasoning
Yongcan Yu, Lingxiao He, Jian Liang +5
Test-time reinforcement learning (TTRL) always adapts models at inference time via pseudo-labeling, leaving it vulnerable to spurious optimization signals from label noise. Through…
Taming Momentum: Rethinking Optimizer States Through Low-Rank Approximation
Zhengbo Wang, Jian Liang, Ran He +2
Modern optimizers like Adam and Muon are central to training large language models, but their reliance on first- and second-order momenta introduces significant memory overhead, wh…
Mitigating the Safety-utility Trade-off in LLM Alignment via Adaptive Safe Context Learning
Yanbo Wang, Minzheng Wang, Jian Liang +3
While reasoning models have achieved remarkable success in complex reasoning tasks, their increasing power necessitates stringent safety measures. For safety alignment, the core ch…
A Comprehensive Survey on Trustworthiness in Reasoning with Large Language Models
Yanbo Wang, Yongcan Yu, Jian Liang +1
The development of Long-CoT reasoning has advanced LLM performance across various tasks, including language understanding, complex problem solving, and code generation. This paradi…
The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models
Lijun Sheng, Jian Liang, Ran He +2
Test-time adaptation (TTA) methods have gained significant attention for enhancing the performance of vision-language models (VLMs) such as CLIP during inference, without requiring…
Test-Time Immunization: A Universal Defense Framework Against Jailbreaks for (Multimodal) Large Language Models
Yongcan Yu, Yanbo Wang, Ran He +1
While (multimodal) large language models (LLMs) have attracted widespread attention due to their exceptional capabilities, they remain vulnerable to jailbreak attacks. Various defe…
Do We Really Need Curated Malicious Data for Safety Alignment in Multi-modal Large Language Models?
Yanbo Wang, Jiyang Guan, Jian Liang +1
Multi-modal large language models (MLLMs) have made significant progress, yet their safety alignment remains limited. Typically, current open-source MLLMs rely on the alignment inh…
R-TPT: Improving Adversarial Robustness of Vision-Language Models through Test-Time Prompt Tuning
Lijun Sheng, Jian Liang, Zilei Wang +1
Vision-language models (VLMs), such as CLIP, have gained significant popularity as foundation models, with numerous fine-tuning methods developed to enhance performance on downstre…
Towards Compatible Fine-tuning for Vision-Language Model Updates
Zhengbo Wang, Jian Liang, Lijun Sheng +3
So far, efficient fine-tuning has become a popular strategy for enhancing the capabilities of foundation models on downstream tasks by learning plug-and-play modules. However, exis…
Sample Correlation for Fingerprinting Deep Face Recognition
Jiyang Guan, Jian Liang, Yanbo Wang +1
Face recognition has witnessed remarkable advancements in recent years, thanks to the development of deep learning techniques.However, an off-the-shelf face recognition model as a…
Prototypical Distillation and Debiased Tuning for Black-box Unsupervised Domain Adaptation
Jian Liang, Lijun Sheng, Hongmin Liu +1
Unsupervised domain adaptation aims to transfer knowledge from a related, label-rich source domain to an unlabeled target domain, thereby circumventing the high costs associated wi…
LoRA-Pro: Are Low-Rank Adapters Properly Optimized?
Zhengbo Wang, Jian Liang, Ran He +2
Low-rank adaptation, also known as LoRA, has emerged as a prominent method for parameter-efficient fine-tuning of foundation models. Despite its computational efficiency, LoRA stil…
A Hard-to-Beat Baseline for Training-free CLIP-based Adaptation
Zhengbo Wang, Jian Liang, Lijun Sheng +3
Contrastive Language-Image Pretraining (CLIP) has gained popularity for its remarkable zero-shot capacity. Recent research has focused on developing efficient fine-tuning methods,…
Towards Eliminating Hard Label Constraints in Gradient Inversion Attacks
Yanbo Wang, Jian Liang, Ran He
Gradient inversion attacks aim to reconstruct local training data from intermediate gradients exposed in the federated learning framework. Despite successful attacks, all previous…
Protecting Model Adaptation from Trojans in the Unlabeled Data
Lijun Sheng, Jian Liang, Ran He +2
Model adaptation tackles the distribution shift problem with a pre-trained model instead of raw data, which has become a popular paradigm due to its great privacy protection. Exist…
Exploring Vacant Classes in Label-Skewed Federated Learning
Kuangpu Guo, Yuhe Ding, Jian Liang +3
Label skews, characterized by disparities in local label distribution across clients, pose a significant challenge in federated learning. As minority classes suffer from worse accu…