Large language models (LLMs) often struggle with mathematical problems that require exact computation or multi-step algebraic reasoning. Tool-integrated reasoning (TIR) offers a promising solution by leveraging external tools such as code interpreters to ensure correctness, but it introduces inference-time dependencies that hinder scalability and deployment. In this work, we propose a new paradigm for distilling tool knowledge into LLMs purely through natural language. We first construct a Solver Agent that solves math problems by interleaving planning, symbolic tool calls, and reflective reasoning. Then, using a back-translation pipeline powered by multiple LLM-based agents, we convert interleaved TIR traces into natural language reasoning traces. A Translator Agent generates explanations for individual tool calls, while a Rephrase Agent merges them into a fluent and globally coherent narrative. Empirically, we show that fine-tuning a small open-source model on these synthesized traces enables it to internalize both tool knowledge and structured reasoning patterns, yielding gains on competition-level math benchmarks without requiring tool access at inference.

The recent work by Shojaee et al. (2025), titled The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity, presents a compelling empirical finding, a reasoning cliff, where the performance of Large Reasoning Models (LRMs) collapses beyond a specific complexity threshold, which the authors posit as an intrinsic scaling limitation of Chain-of-Thought (CoT) reasoning. This commentary, while acknowledging the study's methodological rigor, contends that this conclusion is confounded by experimental artifacts. We argue that the observed failure is not evidence of a fundamental cognitive boundary, but rather a predictable outcome of system-level constraints in the static, text-only evaluation paradigm, including tool use restrictions, context window recall issues, the absence of crucial cognitive baselines, inadequate statistical reporting, and output generation limits. We reframe this performance collapse through the lens of an agentic gap, asserting that the models are not failing at reasoning, but at execution within a profoundly restrictive interface. We empirically substantiate this critique by demonstrating a striking reversal. A model, initially declaring a puzzle impossible when confined to text-only generation, now employs agentic tools to not only solve it but also master variations of complexity far beyond the reasoning cliff it previously failed to surmount. Additionally, our empirical analysis of tool-enabled models like o4-mini and GPT-4o reveals a hierarchy of agentic reasoning, from simple procedural execution to complex meta-cognitive self-correction, which has significant implications for how we define and measure machine intelligence. The illusion of thinking attributed to LRMs is less a reasoning deficit and more a consequence of an otherwise capable mind lacking the tools for action.

Reinforcement Learning (RL)-based post-training has significantly advanced the complex reasoning capabilities of language models, fostering sophisticated self-reflection processes. However, this ``slow thinking'' paradigm presents a critical challenge to reasoning efficiency: models may expend excessive computation on simple questions and shift reasoning prematurely for complex ones. Previous mechanisms typically rely on static length budgets or predefined rules, lacking the adaptability for varying question complexities and models' evolving capabilities. To this end, we propose AdapThink, an adaptive post-training framework designed to induce more efficient thinking while maintaining the performance of reasoning language models. Specifically, AdapThink incorporates two key mechanisms: 1) A group-relative reward function that leverages model confidence and response's characteristic to dynamically adjust the preference of reflection-related transition words without resorting to a fixed length preference. 2) A diversity-aware sampling mechanism that balances the training group's solution accuracy with reasoning diversity via an entropy-guided score. Experiments on several mathematical reasoning datasets with DeepSeek-distilled models demonstrate AdapThink's advantages in enabling adaptive reasoning patterns and mitigating the inefficiencies.

Large language models excel at generating fluent text but frequently struggle with structured reasoning involving temporal constraints, causal relationships, and probabilistic reasoning. To address these limitations, we propose Temporal Causal Probabilistic Description Logic (T-CPDL), an integrated framework that extends traditional Description Logic with temporal interval operators, explicit causal relationships, and probabilistic annotations. We present two distinct variants of T-CPDL: one capturing qualitative temporal relationships through Allen's interval algebra, and another variant enriched with explicit timestamped causal assertions. Both variants share a unified logical structure, enabling complex reasoning tasks ranging from simple temporal ordering to nuanced probabilistic causation. Empirical evaluations on temporal reasoning and causal inference benchmarks confirm that T-CPDL substantially improves inference accuracy, interpretability, and confidence calibration of language model outputs. By delivering transparent reasoning paths and fine-grained temporal and causal semantics, T-CPDL significantly enhances the capability of language models to support robust, explainable, and trustworthy decision-making. This work also lays the groundwork for developing advanced Logic-Retrieval-Augmented Generation (Logic-RAG) frameworks, potentially boosting the reasoning capabilities and efficiency of knowledge graph-enhanced RAG systems.

This paper outlines a comprehensive theoretical and architectural framework for constructing epistemically grounded artificial intelligence systems capable of propositional commitment, metacognitive reasoning, contradiction detection, and normative truth maintenance. Moving beyond the constraints of stochastic language generation, we propose a model in which artificial agents engage in structured, rule-governed reasoning that adheres to explicit epistemic norms. The approach integrates insights from epistemology, formal logic, inferential semantics, knowledge graph structuring, probabilistic justification, and immutable blockchain evidence to create systems that do not merely simulate knowledge, but operate under explicit, verifiable constraints on belief, justification, and truth. We begin with an analysis of epistemic norms in artificial reasoning, contrasting evi-dentialist, Bayesian, and logical foundations, and establishing a requirement for internal consistency and constraint against falsehood. Central to the proposed system is a prohibition against internal deception: no model component may assert what it internally contradicts.

Recently, Agentic AI has become an increasingly popular research field. However, we argue that current agent research practices lack standardization and scientific rigor, making it hard to conduct fair comparisons among methods. As a result, it is still unclear how different design choices in agent frameworks affect effectiveness, and measuring their progress remains challenging. In this work, we conduct a systematic empirical study on GAIA benchmark and BrowseComp to examine the impact of popular design choices in key agent components in a fair and rigorous manner. We find that the lack of a standard evaluation protocol makes previous works, even open-sourced ones, non-reproducible, with significant variance between random runs. Therefore, we introduce a more robust evaluation protocol to stabilize comparisons. Our study reveals which components and designs are crucial for effective agents, while others are redundant, despite seeming logical. Based on our findings, we build and open-source OAgents, a new foundation agent framework that achieves state-of-the-art performance among open-source projects. OAgents offers a modular design for various agent components, promoting future research in Agentic AI.

Recent advancements in large language models (LLMs) have demonstrated substantial progress in reasoning capabilities, such as DeepSeek-R1, which leverages rule-based reinforcement learning to enhance logical reasoning significantly. However, extending these achievements to multimodal large language models (MLLMs) presents critical challenges, which are frequently more pronounced for Multimodal Small Language Models (MSLMs) given their typically weaker foundational reasoning abilities: (1) the scarcity of high-quality multimodal reasoning datasets, (2) the degradation of reasoning capabilities due to the integration of visual processing, and (3) the risk that direct application of reinforcement learning may produce complex yet incorrect reasoning processes. To address these challenges, we design a novel framework Infi-MMR to systematically unlock the reasoning potential of MSLMs through a curriculum of three carefully structured phases and propose our multimodal reasoning model Infi-MMR-3B. The first phase, Foundational Reasoning Activation, leverages high-quality textual reasoning datasets to activate and strengthen the model's logical reasoning capabilities. The second phase, Cross-Modal Reasoning Adaptation, utilizes caption-augmented multimodal data to facilitate the progressive transfer of reasoning skills to multimodal contexts. The third phase, Multimodal Reasoning Enhancement, employs curated, caption-free multimodal data to mitigate linguistic biases and promote robust cross-modal reasoning. Infi-MMR-3B achieves both state-of-the-art multimodal math reasoning ability (43.68% on MathVerse testmini, 27.04% on MathVision test, and 21.33% on OlympiadBench) and general reasoning ability (67.2% on MathVista testmini). Resources are available at https://huggingface.co/Reallm-Labs/Infi-MMR-3B.

Recent advances in large reasoning models (LRMs) demonstrate that sophisticated behaviors such as multi-step reasoning and self-reflection can emerge via reinforcement learning with verifiable rewards~(\textit{RLVR}). However, existing \textit{RLVR} approaches are inherently ``on-policy'', limiting learning to a model's own outputs and failing to acquire reasoning abilities beyond its initial capabilities. To address this issue, we introduce \textbf{LUFFY} (\textbf{L}earning to reason \textbf{U}nder o\textbf{FF}-polic\textbf{Y} guidance), a framework that augments \textit{RLVR} with off-policy reasoning traces. LUFFY dynamically balances imitation and exploration by combining off-policy demonstrations with on-policy rollouts during training. Specifically, LUFFY combines the Mixed-Policy GRPO framework, which has a theoretically guaranteed convergence rate, alongside policy shaping via regularized importance sampling to avoid superficial and rigid imitation during mixed-policy training. Compared with previous RLVR methods, LUFFY achieves an over \textbf{+6.4} average gain across six math benchmarks and an advantage of over \textbf{+6.2} points in out-of-distribution tasks. Most significantly, we show that LUFFY successfully trains weak models in scenarios where on-policy RLVR completely fails. These results provide compelling evidence that LUFFY transcends the fundamental limitations of on-policy RLVR and demonstrates the great potential of utilizing off-policy guidance in RLVR.

We present the Multi-Agent Transformer World Model (MATWM), a novel transformer-based world model designed for multi-agent reinforcement learning in both vector- and image-based environments. MATWM combines a decentralized imagination framework with a semi-centralized critic and a teammate prediction module, enabling agents to model and anticipate the behavior of others under partial observability. To address non-stationarity, we incorporate a prioritized replay mechanism that trains the world model on recent experiences, allowing it to adapt to agents' evolving policies. We evaluated MATWM on a broad suite of benchmarks, including the StarCraft Multi-Agent Challenge, PettingZoo, and MeltingPot. MATWM achieves state-of-the-art performance, outperforming both model-free and prior world model approaches, while demonstrating strong sample efficiency, achieving near-optimal performance in as few as 50K environment interactions. Ablation studies confirm the impact of each component, with substantial gains in coordination-heavy tasks.

Large vision-language models (VLMs) for autonomous driving (AD) are evolving beyond perception and cognition tasks toward motion planning. However, we identify two critical challenges in this direction: (1) VLMs tend to learn shortcuts by relying heavily on history input information, achieving seemingly strong planning results without genuinely understanding the visual inputs; and (2) the chain-ofthought (COT) reasoning processes are always misaligned with the motion planning outcomes, and how to effectively leverage the complex reasoning capability to enhance planning remains largely underexplored. In this paper, we start from a small-scale domain-specific VLM and propose Drive-R1 designed to bridges the scenario reasoning and motion planning for AD. Drive-R1 first undergoes the supervised finetuning on a elaborate dataset containing both long and short COT data. Drive-R1 is encouraged to reason step-by-step from visual input to final planning decisions. Subsequently, Drive-R1 is trained within a reinforcement learning framework that incentivizes the discovery of reasoning paths that are more informative for planning, guided by rewards based on predicted trajectories and meta actions. Experimental evaluations on the nuScenes and DriveLM-nuScenes benchmarks demonstrate that Drive-R1 achieves superior performance compared to existing state-of-the-art VLMs. We believe that Drive-R1 presents a promising direction for bridging reasoning and planning in AD, offering methodological insights for future research and applications.