Related papers: Targeted Neuron Modulation via Contrastive Pair Se…

Contrastive steering has been shown as a simple and effective method to adjust the generative behavior of LLMs at inference time. It uses examples of prompt responses with and without a trait to identify a direction in an intermediate…

Neural predictors have shown great potential in the evaluation process of neural architecture search (NAS). However, current predictor-based approaches overlook the fact that training a predictor necessitates a considerable number of…

Large language models (LLMs) are typically aligned to refuse harmful instructions through safety fine-tuning. A recent attack, termed abliteration, identifies and suppresses the single latent direction most responsible for refusal behavior,…

Reward models (RMs) play a central role in aligning large language models (LLMs) with human preferences. However, RMs are often sensitive to spurious features such as response length. Existing inference-time approaches for mitigating these…

Recent work has shown that LLMs can sometimes detect when steering vectors are injected into their residual stream and identify the injected concept -- a phenomenon termed "introspective awareness." We investigate the mechanisms underlying…

Refusal refers to the functional behavior enabling safety-aligned language models to reject harmful or unethical prompts. Following the growing scientific interest in mechanistic interpretability, recent work encoded refusal behavior as a…

Masked diffusion language models (MDLMs) generate text via iterative masked-token denoising, enabling mask-parallel decoding and distinct controllability and efficiency tradeoffs from autoregressive LLMs. Yet, efficient representation-level…

Despite the strong reasoning capabilities of recent large language models (LLMs), achieving reliable performance on challenging tasks often requires post-training or computationally expensive sampling strategies, limiting their practical…

Current alignment evaluation mostly measures whether models encode dangerous concepts and whether they refuse harmful requests. Both miss the layer where alignment often operates: routing from concept detection to behavioral policy. We…

Large language models exhibit strong multilingual capabilities, yet significant performance gaps persist between dominant and non-dominant languages. Prior work attributes this gap to imbalances between shared and language-specific neurons…

Generative language models (LMs) offer numerous advantages but may produce inappropriate or harmful outputs due to the harmful knowledge acquired during pre-training. This knowledge often manifests as undesirable correspondences, such as…

Generative models are widely used for unsupervised learning with various applications, including data compression and signal restoration. Training methods for such systems focus on the generality of the network given limited amount of…

Ransomware has become a critical threat to cybersecurity due to its rapid evolution, the necessity for early detection, and growing diversity, posing significant challenges to traditional detection methods. While AI-based approaches had…

Refusal behavior in large language models (LLMs) enables them to decline responding to harmful, unethical, or inappropriate prompts, ensuring alignment with ethical standards. This paper investigates refusal behavior across six LLMs from…

Machine learning algorithms can be fooled by small well-designed adversarial perturbations. This is reminiscent of cellular decision-making where ligands (called antagonists) prevent correct signalling, like in early immune recognition. We…

A binary decision task, like yes-no questions or answer verification, reflects a significant real-world scenario such as where users look for confirmation about the correctness of their decisions on specific issues. In this work, we observe…

Susceptibility of neural networks to adversarial attack prompts serious safety concerns for lane detection efforts, a domain where such models have been widely applied. Recent work on adversarial road patches have successfully induced…

Activation steering offers a computationally efficient mechanism for controlling Large Language Models (LLMs) without fine-tuning. While effectively controlling target traits (e.g., persona), coherency degradation remains a major obstacle…

One of the key steps in Neural Architecture Search (NAS) is to estimate the performance of candidate architectures. Existing methods either directly use the validation performance or learn a predictor to estimate the performance. However,…

Deep learning algorithms have been recently targeted by attackers due to their vulnerability. Several research studies have been conducted to address this issue and build more robust deep learning models. Non-continuous deep models are…