相关论文: In-Context Environments Induce Evaluation-Awarenes…

Trustworthy evaluations of dangerous capabilities are increasingly crucial for determining whether an AI system is safe to deploy. One empirically demonstrated threat is sandbagging - the strategic underperformance on evaluations by AI…

Traditional methods for evaluating the robustness of large language models (LLMs) often rely on standardized benchmarks, which can escalate costs and limit evaluations across varied domains. This paper introduces a novel framework designed…

Large Language Models (LLMs) have demonstrated remarkable problem-solving and basic mathematics abilities. However, their efficacy is highly contingent on the formulation of the prompt. This study endeavors to quantify the influence of…

The performance of large language models (LLMs) is acutely sensitive to the phrasing of prompts, which raises significant concerns about their reliability in real-world scenarios. Existing studies often divide prompts into task-level…

Large Language Models (LLMs) are increasingly vulnerable to a sophisticated form of adversarial prompting known as camouflaged jailbreaking. This method embeds malicious intent within seemingly benign language to evade existing safety…

We study how prompt-level inductive biases influence the cognitive behavior of large language models (LLMs) in instructional dialogue. We introduce a symbolic scaffolding method paired with a short-term memory schema designed to promote…

System prompts provide a lightweight yet powerful mechanism for conditioning large language models (LLMs) at inference time. While prior work has focused on English-only settings, real-world deployments benefit from having a single prompt…

In contested domains, instruction-tuned language models must balance user-alignment pressures against faithfulness to the in-context evidence. To evaluate this tension, we introduce a controlled epistemic-conflict framework grounded in the…

AI systems that output their reasoning in natural language offer an opportunity for safety -- we can \emph{monitor} their chain of thought (CoT) for undesirable reasoning, such as the pursuit of harmful objectives. However, the extent to…

Despite significant advances in alignment techniques, we demonstrate that state-of-the-art language models remain vulnerable to carefully crafted conversational scenarios that can induce various forms of misalignment without explicit…

Pre-trained language models (PLMs) have exhibited remarkable few-shot learning capabilities when provided a few examples in a natural language prompt as demonstrations of test instances, i.e., in-context learning. However, the performance…

Insensitivity to semantically-preserving variations of prompts (paraphrases) is crucial for reliable behavior and real-world deployment of large language models. However, language models exhibit significant performance degradation when…

Safety benchmarks are routinely treated as evidence about how a language model will behave once deployed, but this inference is fragile if behavior depends on whether a prompt looks like an evaluation. We define evaluation-context…

While advancements in the reasoning abilities of LLMs have significantly enhanced their performance in solving mathematical problems, coding tasks, and general puzzles, their effectiveness in accurately adhering to instructions remains…

Large language models (LLMs) have transformed the development of embodied intelligence. By providing a few contextual demonstrations, developers can utilize the extensive internal knowledge of LLMs to effortlessly translate complex tasks…

Large language models are able to learn new tasks in context, where they are provided with instructions and a few annotated examples. However, the effectiveness of in-context learning is dependent on the provided context, and the…

In-Context Learning (ICL) allows Large Language Models (LLMs) to adapt to new tasks with just a few examples, but their predictions often suffer from systematic biases, leading to unstable performance in classification. While calibration…

Large language model (LLM) systems increasingly power everyday AI applications such as chatbots, computer-use assistants, and autonomous robots, where performance often depends on manually well-crafted prompts. LLM-based prompt optimizers…

Recent studies have demonstrated that natural-language prompts can help to leverage the knowledge learned by pre-trained language models for the binary sentence-level sentiment classification task. Specifically, these methods utilize…

As the world becomes increasingly saturated with AI-generated content, disinformation, and algorithmic persuasion, critical thinking - the capacity to evaluate evidence, detect unreliable claims, and exercise independent judgment - is…