Related papers: Distractor Injection Attacks on Large Reasoning Mo…

Current LLM safety research predominantly focuses on mitigating Goal Hijacking, preventing attackers from redirecting a model's high-level objective (e.g., from "summarizing emails" to "phishing users"). In this paper, we argue that this…

Large Language Models (LLMs) are attracting significant research attention due to their instruction-following abilities, allowing users and developers to leverage LLMs for a variety of tasks. However, LLMs are vulnerable to prompt-injection…

Large Reasoning Models (LRMs) have achieved tremendous success with their chain-of-thought (CoT) reasoning, yet also face safety issues similar to those of basic language models. In particular, while algorithms are designed to guide them to…

Large Language Models (LLMs) are increasingly described as possessing strong reasoning capabilities, supported by high performance on mathematical, logical, and planning benchmarks. However, most existing evaluations rely on aggregate…

Large Language Models (LLMs) have been found to struggle with systematic reasoning. Even on tasks where they appear to perform well, their performance often depends on shortcuts, rather than on genuine reasoning abilities, leading them to…

Most flagship language models generate explicit reasoning chains, enabling inference-time scaling. However, producing these reasoning chains increases token usage (i.e., reasoning tokens), which in turn increases latency and costs. Our…

As large language models (LLMs) grow more capable, concerns about their safe deployment have also grown. Although alignment mechanisms have been introduced to deter misuse, they remain vulnerable to carefully designed adversarial prompts.…

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,…

The widespread adoption of thinking mode in large language models (LLMs) has significantly enhanced complex task processing capabilities while introducing new security risks. When subjected to jailbreak attacks, the step-by-step reasoning…

Despite rapid advancements in large language models (LLMs), the token-level autoregressive nature constrains their complex reasoning capabilities. To enhance LLM reasoning, inference-time techniques, including…

Long-context large language models (LLMs) are prone to be distracted by irrelevant contexts. The reason for distraction remains poorly understood. In this paper, we first identify the contextual heads, a special group of attention heads…

LLMs demonstrate performance comparable to human abilities in complex tasks such as mathematical reasoning, but their robustness in mathematical reasoning under minor input perturbations still lacks systematic investigation. Existing…

Recent advances in large reasoning models (LRMs) have enabled strong chain-of-thought (CoT) generation through test-time computation. While these multi-step reasoning capabilities represent a major milestone in language model performance,…

Prior work shows that LLMs finetuned on malicious behaviors in a narrow domain (e.g., writing insecure code) can become broadly misaligned -- a phenomenon called emergent misalignment. We investigate whether this extends from conventional…

While tool learning significantly enhances the capabilities of large language models (LLMs), it also introduces substantial security risks. Prior research has revealed various vulnerabilities in traditional LLMs during tool learning.…

Large Reasoning Models (LRMs) represent a breakthrough in AI problem-solving capabilities, but their effectiveness in interactive environments can be limited. This paper introduces and analyzes overthinking in LRMs. A phenomenon where…

Large language models (LLMs) have gained widespread recognition for their superior comprehension and have been deployed across numerous domains. Building on Chain-of-Thought (CoT) ideology, Large Reasoning models (LRMs) further exhibit…

The emergence of Multimodal Large Language Models (MLRMs) has enabled sophisticated visual reasoning capabilities by integrating reinforcement learning and Chain-of-Thought (CoT) supervision. However, while these enhanced reasoning…

Large Reasoning Models (LRMs) leverage Chain-of-Thought (CoT) reasoning to solve complex tasks, but this explicit reasoning process introduces a critical vulnerability: adversarial manipulation of the thought chain itself, known as…

Multimodal LLMs (MLLMs) with a great ability of text and image understanding have received great attention. To achieve better reasoning with MLLMs, Chain-of-Thought (CoT) reasoning has been widely explored, which further promotes MLLMs'…