Search: “reasoning”

arXiv:2609.15987v1 Announce Type: cross Abstract: Reinforcement learning with verifiable rewards (RLVR) improves the reasoning capabilities of large language models (LLMs). We introduce Bellman Policy

arXiv:2609.09565v1 Announce Type: new Abstract: Agentic graph learning (AGL) has recently achieved promising results on graph reasoning tasks, where an agent powered by a large language model (LLM) se

arXiv:2609.07022v1 Announce Type: cross Abstract: Reasoning over privacy-constrained records requires combining structured attributes with evidence from free-text narratives. We introduce CIPHER (Cros

arXiv:2609.09300v1 Announce Type: new Abstract: Video understanding demands a convergence of complementary capabilities across perception, temporal understanding, and complex reasoning, which are diff

arXiv:2609.05518v1 Announce Type: new Abstract: Despite the strong capabilities of multimodal large language models (MLLMs), their parametric knowledge remains incomplete and difficult to update, moti

I was running into a lot of posts that praised both the Fixed template and the Sharp template in comparison to the stock one, so I put them to the test. It's not as extensive as it should be for a pap

arXiv:2609.03941v1 Announce Type: cross Abstract: RL-based post-training for reasoning models is increasingly bottlenecked by repeated fresh rollout generation, particularly in agentic settings where

arXiv:2609.02438v1 Announce Type: new Abstract: Large language models can look capable of logical reasoning, but correct or incorrect answers alone tell us little about what the model represents inter

common, server : enable preserve_reasoning kwarg by default, log its effective state (#28174) common, server : enable preserve_reasoning kwarg by default, log its effective state If the preserve_reaso

arXiv:2608.30463v1 Announce Type: new Abstract: Solvability detection is one of the most challenging aspects of mathematical reasoning for Large Language Models (LLMs). While prior work has studied th

arXiv:2608.30528v1 Announce Type: new Abstract: Reinforcement learning (RL) is used to improve the reasoning abilities of LLMs, while training data span heterogeneous tasks. However, most RL post-trai

arXiv:2608.30270v1 Announce Type: new Abstract: Indirect speech acts (ISAs) require pragmatic reasoning over context, as directive intent can- not be inferred from surface form alone. Prior text-based

arXiv:2608.26136v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) decompose language-model activations into sparse, interpretable features, and an appealing way to aim them at reasoning is to

arXiv:2601.17027v2 Announce Type: replace-cross Abstract: While synthetic data has proven effective for improving scientific reasoning in the text domain, multimodal reasoning remains constrained by t

TL;DR - A ~4M example, ~3B token CoT dataset designed around helping small models think more concisely, accurately and reliably. Hi all! For the past few months I have been working on a dataset design

arXiv:2608.20237v1 Announce Type: new Abstract: Multimodal large language models (MLLMs) combine linguistic reasoning with visual perception, yet their ability to perform visual spatial planning under

arXiv:2608.13958v1 Announce Type: new Abstract: How do we govern AI systems whose reasoning we cannot fully inspect? Governance does not require understanding a system's reasoning. It requires stating

Yes it uses a ton more reasoning tokens than 3.6 did But test in on the same tasks with the other chinese models, glm 5.3, deepseek v4 flash and pro, etc it's really similar, and they are needed The r

arXiv:2608.02820v1 Announce Type: cross Abstract: Chain-of-thought (CoT) monitoring is an increasingly important component of AI safety stacks but relies on the assumption that a model's reasoning tra

arXiv:2608.03745v1 Announce Type: new Abstract: Chain-of-Thought (CoT) reasoning offers a promising window into model monitoring. However, monitoring relies on faithfulness, i.e., the model output str

arXiv:2601.02854v2 Announce Type: replace Abstract: As an agent-level reasoning and coordination paradigm, Multi-Agent Debate (MAD) orchestrates multiple agents through structured debate to improve an

arXiv:2607.27610v1 Announce Type: new Abstract: Reinforcement learning (RL) finetuning significantly enhances the reasoning capabilities of large language models (LLMs), yet its effectiveness critical

arXiv:2607.25182v1 Announce Type: cross Abstract: The ability to retrieve relevant tables for answering questions is a key task for structured information retrieval. Multi-stage retrieval systems rely

arXiv:2607.18985v2 Announce Type: replace Abstract: Large language models (LLMs) have demonstrated remarkable capabilities in language understanding, reasoning, and world knowledge. As embodied agents

arXiv:2607.22609v1 Announce Type: new Abstract: Large Language Models (LLMs) are increasingly used for decision-making and reasoning tasks, yet their potential as controllers for physical systems rema

The math behind reinforcement learning (RL) post-training for large language models (LLMs) is notoriously unforgiving. As frontier AI labs push the boundaries of reasoning and coding models using RL p

arXiv:2511.09008v2 Announce Type: replace-cross Abstract: Large Language Models perform well at natural language interpretation and reasoning, but their lack of formal correctness guarantees limits th

arXiv:2606.31543v1 Announce Type: new Abstract: Large language models can produce fluent, internally coherent reasoning traces for abstract reasoning tasks while still being confidently wrong - making

arXiv:2511.14900v2 Announce Type: replace-cross Abstract: Vision--language models (VLMs) have recently shown promise for assisting clinical reasoning in dermatological diagnosis. However, their trustw

arXiv:2606.26535v1 Announce Type: cross Abstract: Current VLM evaluations often conflate language priors with genuine spatial reasoning. To address this, we introduce CRISP, a novel structural-diagnos

arXiv:2511.05933v2 Announce Type: replace Abstract: Reinforcement learning (RL) is often credited with improving language model reasoning at the expense of knowledge. We challenge this narrative by sh

arXiv:2606.10296v1 Announce Type: cross Abstract: Multi-agent debate systems are typically evaluated only on whether the final answer is correct, overlooking the quality of the intermediate reasoning

arXiv:2606.07157v1 Announce Type: new Abstract: Many efforts to ensure frontier AI models are safe rely on monitoring their chain-of-thought (CoT) reasoning. If models become able to perform sufficien

arXiv:2509.04027v3 Announce Type: replace-cross Abstract: Test-time scaling, primarily manifested through multi-step Chain-of-Thought (CoT) reasoning via Reinforcement Learning (RL), has emerged as a

arXiv:2604.03789v2 Announce Type: replace-cross Abstract: Recent advances in large language models have significantly improved their ability to perform mathematical reasoning, extending from elementar

arXiv:2606.00869v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) has become central to LLM reasoning, but its outcome-level rewards can make models more willing to

arXiv:2605.31062v1 Announce Type: new Abstract: Large Language Models (LLMs) have achieved remarkable performance in complex reasoning tasks through Chain-of-Thought (CoT) prompting. However, this app

arXiv:2601.07525v2 Announce Type: replace-cross Abstract: Natural generation allows Large Language Models (LLMs) to produce free-form responses with rich reasoning, yet the lack of structure makes out

arXiv:2605.28842v1 Announce Type: cross Abstract: The success of large language models (LLMs) across diverse NLP tasks has elevated the importance of reasoning chain optimization as a critical step in

arXiv:2505.13775v4 Announce Type: replace-cross Abstract: Recent impressive results from large reasoning models have been interpreted as a triumph of Chain of Thought (CoT), and especially of the proc

arXiv:2605.25813v1 Announce Type: new Abstract: Embodied Question Answering (EQA) connects perception, reasoning, and interaction within embodied environments. However, existing datasets and benchmark

arXiv:2605.14071v1 Announce Type: new Abstract: Distilling reasoning traces from strong large language models into smaller ones is a promising route to improve intelligence in resource-constrained set

arXiv:2605.07721v1 Announce Type: cross Abstract: Recurrent LLM architectures have emerged as a promising approach for improving reasoning, as they enable multi-step computation in the embedding space

arXiv:2510.24941v3 Announce Type: replace Abstract: Large language models can generate long chain-of-thought (CoT) reasoning, but it remains unclear whether the verbalized steps reflect the models' in

arXiv:2604.20983v1 Announce Type: cross Abstract: Vision evaluations are typically done through multi-step processes. In most contemporary fields, experts analyze images using structured, evidence-bas

arXiv:2601.02989v2 Announce Type: replace Abstract: Large language models (LLMs), despite strong performance on complex mathematical problems, exhibit systematic limitations in counting tasks. This is