Related papers: Refusal in Language Models Is Mediated by a Single…
Multimodal large language models (MLLMs) often fail to transfer safety capabilities learned in the text modality to semantically equivalent non-text inputs, revealing a persistent multimodal safety gap. We study this gap from a…
We find that language models have difficulties generating fallacious and deceptive reasoning. When asked to generate deceptive outputs, language models tend to leak honest counterparts but believe them to be false. Exploiting this…
Large language models (LLMs) achieve impressive performance across diverse tasks yet remain vulnerable to jailbreak attacks that bypass safety mechanisms. We present RAID (Refusal-Aware and Integrated Decoding), a framework that…
When a user tells an AI system that someone "should not" take an action, the system ought to treat this as a prohibition. Yet many large language models do the opposite: they interpret negated instructions as affirmations. We audited 16…
Large Language Models(LLMs) have been successful in numerous fields. Alignment has usually been applied to prevent them from harmful purposes. However, aligned LLMs remain vulnerable to jailbreak attacks that deliberately mislead them into…
Multimodal Large Language Models (MLLMs) have showcased impressive performance in a variety of multimodal tasks. On the other hand, the integration of additional image modality may allow the malicious users to inject harmful content inside…
Inasmuch as the removal of refusal behavior from instruction-tuned language models by directional abliteration requires the extraction of refusal-mediating directions from the residual stream activation space, and inasmuch as the…
We propose affine concept editing (ACE) as an approach for steering language models' behavior by intervening directly in activations. We begin with an affine decomposition of model activation vectors and show that prior methods for steering…
Safety fine-tuning helps align Large Language Models (LLMs) with human preferences for their safe deployment. To better understand the underlying factors that make models safe via safety fine-tuning, we design a synthetic data generation…
In this paper, we present empirical studies on conversational recommendation tasks using representative large language models in a zero-shot setting with three primary contributions. (1) Data: To gain insights into model behavior in…
Multi-turn jailbreaks exploit the ability of large language models to accumulate and act on conversational context. Instead of stating a harmful request directly, an attacker can gradually steer the conversation toward an unsafe answer.…
Multimodal AI systems integrate text generation, image generation, and other capabilities within a single conversational interface. These systems employ safety mechanisms to prevent disallowed actions, including the removal of watermarks…
Representation Engineering analyses often characterize refusal using static directions extracted from terminal or pooled representations. We ask whether this view misses how refusal is constructed across layer-token positions. Using causal…
The deployment of Large Language Models (LLMs) as tool-using agents causes their alignment training to manifest in new ways. Recent work finds that language models can use tools in ways that contradict the interests or explicit instructions…
We present MultiBreak, a scalable and diverse multi-turn jailbreak benchmark to evaluate large language model (LLM) safety. Multi-turn jailbreaks mimic natural conversational settings, making them easier to bypass safety-aligned LLM than…
Large Language Models (LLMs) exhibit strong conversational abilities but often generate falsehoods. Prior work suggests that the truthfulness of simple propositions can be represented as a single linear direction in a model's internal…
Current LLMs are trained to refuse potentially harmful input queries regardless of whether users actually had harmful intents, causing a tradeoff between safety and user experience. Through a study of 480 participants evaluating 3,840…
Pretrained large language models (LLMs) are becoming increasingly powerful and ubiquitous in mainstream applications such as being a personal assistant, a dialogue model, etc. As these models become proficient in deducing user preferences…
Language models can be persuaded to abandon factual knowledge. This vulnerability is central to AI safety, but its internal mechanism remains poorly understood. We uncover a compact causal mechanism for persuasion-induced factual errors. A…
The increased presence of large language models (LLMs) in educational settings has ignited debates concerning negative repercussions, including overreliance and inadequate task reflection. Our work advocates moderated usage of such models,…