相关论文: Detecting and Steering LLMs' Empathy in Action

Recent studies on the safety alignment of large language models (LLMs) have revealed that existing approaches often operate superficially, leaving models vulnerable to various adversarial attacks. Despite their significance, these studies…

Unlearning in Large Language Models (LLMs) is crucial for protecting private data and removing harmful knowledge. Most existing approaches rely on fine-tuning to balance unlearning efficiency with general language capabilities. However,…

Large language models (LLMs) trained on internet-scale corpora can exhibit systematic biases that increase the probability of unwanted behavior. In this study, we examined potential biases towards synthetic vs. biological technological…

Mechanistic interpretability has made it possible to localize circuits underlying specific behaviors in language models, but existing methods are expensive, model-specific, and difficult to scale to larger architectures. We introduce…

Implicit biases refer to automatic mental processes that shape perceptions, judgments, and behaviors. Previous research on "implicit bias" in LLMs focused primarily on outputs rather than the processes underlying the outputs. We present the…

Representation Misdirection for Unlearning (RMU), which steers model representation in the intermediate layer to a target random representation, is an effective method for large language model (LLM) unlearning. Despite its high performance,…

Mitigating reward hacking--where AI systems misbehave due to flaws or misspecifications in their learning objectives--remains a key challenge in constructing capable and aligned models. We show that we can monitor a frontier reasoning…

Large Language Models (LLMs) are increasingly used as high level controllers for autonomous Unmanned Aerial Vehicle (UAV) missions. However, existing evaluations rarely assess whether such agents remain safe, protocol compliant, and…

Large Language Models (LLMs) are integral to modern AI applications, but their safety alignment mechanisms can be bypassed through adversarial prompt engineering. This study investigates emoji-based jailbreaking, where emoji sequences are…

When multiple large language models interact in a shared conversation, do they develop differentiated social roles or converge toward uniform behavior? We present a controlled experimental platform that orchestrates simultaneous multi-agent…

LLM agents do not act on raw interaction history; they act on a bounded decision state assembled by truncation, summarization, reordering, and rewriting. If directive-bearing state is dropped, weakened, or rebound during that step, an agent…

As autonomous agents become more capable of performing real-world tasks, distinguishing scheming behavior from benign task pursuit may become a central AI control problem. Existing monitors often rely on chain-of-thought access or internal…

Large language models demonstrate strong performance on mathematical reasoning benchmarks, yet remain surprisingly fragile to meaning-preserving surface perturbations. We systematically evaluate three open-weight LLMs, Mistral-7B,…

Large Language Models (LLMs) often generate inconsistent responses when prompted with semantically equivalent paraphrased inputs. Recently, activation steering, a technique that modulates LLMs' behaviours by adjusting their latent…

An OS kernel that runs LLM inference internally can read logit distributions before any text is generated and act on them as a governance primitive. This paper presents ProbeLogits, a kernel-level operation that performs a single forward…

Evolutionary Robotics allows robots with limited sensors and processing to tackle complex tasks by means of sensory-motor coordination. In this paper we show the first application of the Behaviour Tree framework to a real robotic platform…

Deploying LLMs raises two coupled challenges: (1) monitoring--estimating where a model underperforms as traffic and domains drift--and (2) improvement--prioritizing data acquisition to close the largest performance gaps. We test whether an…

Large language models (LLMs) can internally distinguish between evaluation and deployment contexts, a behaviour known as \emph{evaluation awareness}. This undermines AI safety evaluations, as models may conceal dangerous capabilities during…

Autonomous driving systems often infer pedestrian yielding behavior from geometric and kinematic cues alone, limiting their ability to reason about visual scene context and age-dependent behavioral variability. This limitation can produce…

Artificially intelligent (AI) co-scientists must be able to sift through research literature cost-efficiently while applying nuanced scientific reasoning. We evaluate Small Language Models (SLMs, <= 8B parameters) for classifying medical…