相关论文: InvThink: Premortem Reasoning for Safer Language M…

Due to their architecture and vast pre-training data, large language models (LLMs) demonstrate strong text classification performance. However, LLM output - here, the category assigned to a text - depends heavily on the wording of the…

While Multimodal Large Language Models (MLLMs) have made remarkable progress in vision-language reasoning, they are also more susceptible to producing harmful content compared to models that focus solely on text. Existing defensive…

Large Reasoning Models (LRMs) leverage transparent reasoning traces, known as Chain-of-Thoughts (CoTs), to break down complex problems into intermediate steps and derive final answers. However, these reasoning traces introduce unique safety…

Large reasoning models achieve strong performance on complex tasks by generating extended chains of thought, but they often "overthink": continuing to reason long after they have enough information to answer correctly. This wastes…

Ensuring safe and appropriate responses from vision-language models (VLMs) remains a critical challenge, particularly in high-risk or ambiguous scenarios. We introduce SafeCoT, a lightweight, interpretable framework that leverages…

Recent years have witnessed significant progress in large language models' (LLMs) reasoning, which is largely due to the chain-of-thought (CoT) approaches, allowing models to generate intermediate reasoning steps before reaching the final…

Preference alignment has enabled large language models (LLMs) to better reflect human expectations, but current methods mostly optimize for population-level preferences, overlooking individual users. Personalization is essential, yet early…

Large Vision-Language Models (LVLMs) have achieved remarkable progress in multimodal perception and generation, yet their safety alignment remains a critical challenge.Existing defenses and vulnerable to multimodal jailbreaks, as visual…

Humans ponder before articulating complex sentence elements, enabling deeper cognitive processing through focused effort. In this work, we introduce this pondering process into language models by repeatedly invoking the forward process…

Inductive reasoning is a core problem-solving capacity: humans can identify underlying principles from a few examples, which robustly generalize to novel scenarios. Recent work evaluates large language models (LLMs) on inductive reasoning…

Solving complex, long-horizon robotic manipulation tasks requires a deep understanding of physical interactions, reasoning about their long-term consequences, and precise high-level planning. Vision-Language Models (VLMs) offer a general…

Safety alignment in large language models (LLMs) is achieved through fine-tuning mechanisms that regulate neuron activations to suppress harmful content. In this work, we propose a novel approach to induce disalignment by identifying and…

Reasoning is an important task for large language models (LLMs). Among all the reasoning paradigms, inductive reasoning is one of the fundamental types, which is characterized by its particular-to-general thinking process and the…

We discover a novel and surprising phenomenon of unintentional misalignment in reasoning language models (RLMs), which we call self-jailbreaking. Specifically, after benign reasoning training on math or code domains, RLMs will use multiple…

With the rise of advanced reasoning capabilities, large language models (LLMs) are receiving increasing attention. However, although reasoning improves LLMs' performance on downstream tasks, it also introduces new security risks, as…

Recently, large language models (LLMs) have demonstrated outstanding reasoning capabilities on mathematical and coding tasks. However, their application to financial tasks-especially the most fundamental task of stock movement…

We explore how generating a chain of thought -- a series of intermediate reasoning steps -- significantly improves the ability of large language models to perform complex reasoning. In particular, we show how such reasoning abilities emerge…

Small language models (SLMs) often struggle with complex mathematical reasoning due to limited capacity to maintain long chains of intermediate steps and to recover from early errors. We address this challenge by introducing a hint-assisted…

Recent advancements in large language models (LLMs) have catalyzed the rise of reasoning-intensive inference paradigms, where models perform explicit step-by-step reasoning before generating final answers. While such approaches improve…

Vision-language models (VLMs) demonstrate strong multimodal capabilities but have been found to be more susceptible to generating harmful content compared to their backbone large language models (LLMs). Our investigation reveals that the…