Related papers: Monitoring Reasoning Models for Misbehavior and th…

The integration of slow-thinking mechanisms into large language models (LLMs) offers a promising way toward achieving Level 2 AGI Reasoners, as exemplified by systems like OpenAI's o1. However, several significant challenges remain,…

With the rapid rise of personalized AI, customized large language models (LLMs) equipped with Chain of Thought (COT) reasoning now power millions of AI agents. However, their complex reasoning processes introduce new and largely unexplored…

Large Language Models (LLMs) are pivotal in advancing natural language processing but often struggle with complex reasoning tasks due to inefficient attention distributions. In this paper, we explore the effect of increased computed tokens…

Reasoning models excel at complex tasks such as coding and mathematics, yet their inference is often slow and token-inefficient. To improve the inference efficiency, post-training quantization (PTQ) usually comes with the cost of large…

Chain-of-thought (CoT) reasoning has become the standard paradigm for enabling Large Language Models (LLMs) to solve complex problems. However, recent studies reveal a sharp performance drop in reasoning hop generalization scenarios, where…

Chain of thought (CoT) fine-tuning aims to endow large language models (LLMs) with reasoning capabilities by training them on curated reasoning traces. It leverages both supervised and reinforced fine-tuning to cultivate human-like…

OpenAI o1 has shown that applying reinforcement learning to integrate reasoning steps directly during inference can significantly improve a model's reasoning capabilities. This result is exciting as the field transitions from the…

Chain of Thought (CoT) reasoning enhances language models' performance but often leads to inefficient "overthinking" on simple problems. We identify that existing approaches directly penalizing reasoning length fail to account for varying…

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…

Chain-of-thought (CoT) reasoning in vision language models (VLMs) is crucial for improving interpretability and trustworthiness. However, current training recipes lack robust CoT reasoning data, relying on datasets dominated by short…

Recent advances in large multimodal models have leveraged image-based tools with reinforcement learning to tackle visual problems. However, existing open-source approaches often exhibit monotonous reasoning patterns and allow only a limited…

We study fair multi-agent multi-armed bandit learning under collision-only coordination. Agents cannot communicate explicitly during learning and observe only their own rewards and whether collisions occur when several agents access the…

Chain-of-Thought (CoT) prompting has emerged as a foundational technique for eliciting reasoning from Large Language Models (LLMs), yet the robustness of this approach to corruptions in intermediate reasoning steps remains poorly…

Large language models can generate long chain-of-thought (CoT) reasoning, yet prior work suggests that CoT can be post-hoc rationalization rather than a faithful reflection of the computation through explicitly designed settings. In this…

Recent work has highlighted the importance of monitoring chain-of-thought reasoning for AI safety; however, current approaches that analyze textual reasoning steps can miss subtle harmful patterns and may be circumvented by models that hide…

In this work, we examine how targeted perturbations in the activation space of Language Models (LMs) can encode complex reasoning patterns. We inject steering vectors, derived from LM activations, into LMs during inference time and study…

In information retrieval, large language models (LLMs) have demonstrated remarkable potential in text reranking tasks by leveraging their sophisticated natural language understanding and advanced reasoning capabilities. However,…

Affordances and permissions are promising and timely safety levers for mitigating Loss of Control (LoC) threats in high-stakes deployment contexts, such as national security. Deployers in defense and intelligence could rely on several…

Large Reasoning Models (LRMs) have demonstrated remarkable capabilities in complex problem-solving through Chain-of-Thought (CoT) reasoning. However, the multi-step nature of CoT introduces new safety challenges that extend beyond…

Latent tokens are gaining attention for enhancing reasoning in large language models (LLMs), yet their internal mechanisms remain unclear. This paper examines the problem from a reliability perspective, uncovering fundamental weaknesses:…