15 accepted papers
While Large Language Models (LLMs) have achieved remarkable success in dyadic (one-on-one) instruction, they face significant challenges in One-to-Many alignment, such as clinical ward rounds, where an instructor must simultaneously guide a diverse group of trainees. Current models often suffer from…
Representation intervention aims to locate and modify the representations that encode the underlying concepts in Large Language Models (LLMs) to elicit the aligned and expected behaviors. Despite the empirical success, it has never been examined whether one could locate the faithful concepts for int…
Large Language Models (LLMs) excel at long-context understanding but exhibit significant limitations in long-form generation. Existing studies primarily focus on single-generation quality, generally overlooking the volatility of the output (i.e., the inconsistency in length and content across multip…
Efficient distillation is a key pathway for converting expensive reasoning capability into deployable efficiency, yet in the frontier regime where the student already has strong reasoning ability, naive continual distillation often yields limited gains or even degradation. We observe a characteristi…
Autonomous agents have recently achieved remarkable progress across diverse domains, yet most evaluations focus on short-horizon, fully observable tasks. In contrast, many critical real-world tasks, such as large-scale software development, commercial investment, and scientific discovery, unfold in …
LLMs have made impressive progress, but their growing capabilities also expose them to highly flexible jailbreaking attacks designed to bypass safety alignment. While many existing defenses focus on known types of attacks, it is more critical to prepare LLMs for *unseen* attacks that may arise durin…
Large Language Models (LLMs) typically rely on Supervised Fine-Tuning (SFT) to specialize in downstream tasks, with the Cross Entropy (CE) loss being the de facto choice. However, CE maximizes the likelihood of observed data without accounting for alternative possibilities. As such, CE usually lead…
Preserving intellectual property (IP) within a pre-trained diffusion model is critical for protecting the model's copyright and preventing unauthorized model deployment. In this regard, model watermarking is a common practice for IP protection that embeds traceable information within models and allo…
Training safe LLMs remains a critical challenge. The most widely used method, Refusal Training (RT), struggles to generalize against various Out-of-Distribution (OOD) jailbreaking attacks. Although various advanced methods have been proposed to address this issue, we instead question whether OOD att…
Backdoor attacks pose a significant threat to Deep Neural Networks (DNNs) as they allow attackers to manipulate model predictions with backdoor triggers. To address these security vulnerabilities, various backdoor purification methods have been proposed to purify compromised models. Typically, these…
Large language models (LLMs) outperform information retrieval techniques for downstream knowledge-intensive tasks when being prompted to generate world knowledge. However, community concerns abound regarding the factuality and potential implications of using this uncensored knowledge. In light of th…
Imitation learning (IL) algorithms excel in acquiring high-quality policies from expert data for sequential decision-making tasks. But, their effectiveness is hampered when faced with limited expert data. To tackle this challenge, a novel framework called (offline) IL with supplementary data has bee…
It has been widely observed that deep neural networks (DNN) are vulnerable to backdoor attacks where attackers could manipulate the model behavior maliciously by tampering with a small set of training samples. Although a line of defense methods is proposed to mitigate this threat, they either requir…
Deep neural networks (DNNs) have been shown to be vulnerable to adversarial examples, which can produce erroneous predictions by injecting imperceptible perturbations. In this work, we study the transferability of adversarial examples, which is significant due to its threat to real-world application…
The query-based black-box attacks have raised serious threats to machine learning models in many real applications. In this work, we study a lightweight defense method, dubbed Random Noise Defense (RND), which adds proper Gaussian noise to each query. We conduct the theoretical analysis about the ef…