研究人员开发了新的框架来增强AI模型的推理能力。一种方法，流动推理模型（FRMs），使用迭代自我完善和动态稳定性检查来高精度地解决数独等复杂谜题。另一种方法，SemFlowRAG，通过创建有向语义梯度图来指导模型从抽象概念到具体证据，避免“概率黑洞”，从而改进检索增强生成。此外，数据高效蒸馏框架（DED）使用精选数据集和最优教师模型，无需大规模扩展即可实现强大的推理性能，为高级AI推理提供了实用途径。
AI
Reasoning Large Language Models (LLMs) enable test-time scaling, with dataset-level accuracy improving as the token budget increases, motivating adaptive reasoning—spending tokens when they improve reliability and stopping early when additional computation is unlikely to help. Ho…
arXiv:2607.01571v1 Announce Type: new Abstract: Chain-of-thought (CoT) reasoning enables large language models (LLMs) to solve complex problems by generating intermediate reasoning steps. While much attention has been paid to the length and content of these reasoning chains, far …
arXiv:2603.02112v2 Announce Type: replace-cross Abstract: Modern language models reason within bounded context, an inherent constraint that poses a fundamental barrier to long-horizon reasoning. We identify recursion as a core principle for overcoming this barrier, and propose re…
arXiv:2607.02262v1 Announce Type: new Abstract: Reasoning Language Models (RLMs) have significantly improved performance on complex tasks by extending the reasoning chain. However, these chains are prone to containing factual errors, particularly in knowledge-intensive tasks. To …
arXiv:2607.01927v1 Announce Type: cross Abstract: This paper presents TUDUM (T\"urk\c{c}e D\"u\c{s}\"unen \"Uretken Model), a project pipeline for adapting a Qwen-family 27B thinking model toward Turkish reasoning. The central problem is not only to answer Turkish prompts in Turk…
arXiv cs.AI
TIER_1English(EN)·Timo Bertram, Sidhant Bhavnani, Richard Freinschlag, Erich Kobler, Andreas Mayr, G\"unter Klambauer·
arXiv:2607.02491v1 Announce Type: new Abstract: In this work, we focus on SE-RRMs, a symbol-equivariant instantiation of RRMs that exhibits improved extrapolation to larger problem sizes. We propose a neuro-symbolic approach, ``Guiding with Recurrent Reasoning Models'' (G-RRM), w…
arXiv cs.AI
TIER_1English(EN)·Ya Gao, Pekka Marttinen·
arXiv:2607.02073v1 Announce Type: new Abstract: Long-context reasoning requires models to locate, revise, and synthesize evidence distributed across lengthy inputs. Existing long-context RL methods usually reward final answers or static evidence extraction, offering little feedba…
In this work, we focus on SE-RRMs, a symbol-equivariant instantiation of RRMs that exhibits improved extrapolation to larger problem sizes. We propose a neuro-symbolic approach, ``Guiding with Recurrent Reasoning Models'' (G-RRM), which integrates SE-RRMs with symbolic solvers fo…
Reasoning Language Models (RLMs) have significantly improved performance on complex tasks by extending the reasoning chain. However, these chains are prone to containing factual errors, particularly in knowledge-intensive tasks. To address this issue, we propose CheckRLM, a frame…
Reasoning Language Models (RLMs) have significantly improved performance on complex tasks by extending the reasoning chain. However, these chains are prone to containing factual errors, particularly in knowledge-intensive tasks. To address this issue, we propose CheckRLM, a frame…
Long-context reasoning requires models to locate, revise, and synthesize evidence distributed across lengthy inputs. Existing long-context RL methods usually reward final answers or static evidence extraction, offering little feedback on how intermediate actions change the model'…
This paper presents TUDUM (Türkçe Düşünen Üretken Model), a project pipeline for adapting a Qwen-family 27B thinking model toward Turkish reasoning. The central problem is not only to answer Turkish prompts in Turkish, but to make the explicit reasoning trace itself Turkish. A th…
arXiv cs.CL
TIER_1English(EN)·Xuecheng Liu, Daman Arora, Gokul Swamy, Andrea Zanette·
arXiv:2607.01077v1 Announce Type: new Abstract: While inference-time scaling has improved the reasoning abilities of large language models (LLMs), the need to generate long chains-of-thought (CoTs) is a computational bottleneck. Thus, in contrast to sequential scaling methods lik…
arXiv cs.AI
TIER_1English(EN)·Ben Slivinski, Michael Saldivar·
arXiv:2607.01223v1 Announce Type: new Abstract: When should an AI system's answer be trusted? Formal proof assistants offer certainty but cannot reach most of the problem distribution; scalar LLM judges offer coverage but produce opaque scores that cannot be audited after the fac…
arXiv:2607.00862v1 Announce Type: cross Abstract: Large Reasoning Models (LRMs) have achieved remarkable success on complex tasks by leveraging long chain-of-thought (CoT) trajectories, yet they frequently exhibit overthinking on simple queries, resulting in significant token ove…
When should an AI system's answer be trusted? Formal proof assistants offer certainty but cannot reach most of the problem distribution; scalar LLM judges offer coverage but produce opaque scores that cannot be audited after the fact and are subject to the same coherence issues a…
While inference-time scaling has improved the reasoning abilities of large language models (LLMs), the need to generate long chains-of-thought (CoTs) is a computational bottleneck. Thus, in contrast to sequential scaling methods like CoT, recent parallel scaling techniques instea…
Large Reasoning Models (LRMs) have achieved remarkable success on complex tasks by leveraging long chain-of-thought (CoT) trajectories, yet they frequently exhibit overthinking on simple queries, resulting in significant token overhead and reduced inference efficiency. However, e…
arXiv cs.AI
TIER_1English(EN)·Wei-Rui Chen, Vignesh Kothapalli, Ata Fatahibaarzi, Hejian Sang, Shao Tang, Qingquan Song, Zhipeng Wang, Muhammad Abdul-Mageed·
arXiv:2512.21002v3 Announce Type: replace-cross Abstract: Distilling the capabilities from a large reasoning model (LRM) to a smaller student model often involves training on substantial amounts of reasoning data. However, knowledge distillation (KD) over lengthy sequences with p…
arXiv cs.AI
TIER_1English(EN)·Alec Helbling, Andrey Bryutkin, Mauro Martino, Nima Dehmamy, Hendrik Strobelt·
arXiv:2606.29150v1 Announce Type: new Abstract: Discrete flow models have recently shown promising performance on few-step text generation; however, when naively applied to structured reasoning tasks such as Sudoku and Zebra puzzles, they converge confidently to incorrect answers…
arXiv cs.AI
TIER_1English(EN)·Houyuan Qin, Rong Wu, Qinyuan Qin, Botian Shi, Jingjing Qu, Yang Sun, Pinlong Cai·
arXiv:2606.28447v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) enhanced by Knowledge Graphs has shown promise in complex multi-hop reasoning tasks. However, existing graph-based retrieval methods typically rely on flat, undirected topologies. During the re…
arXiv cs.AI
TIER_1English(EN)·Xiaojun Wu, Xiaoguang Jiang, Huiyang Li, Jucai Zhai, Dengfeng Liu, Qiaobo Hao, Huang Liu, Zhiguo Yang, Ji Xie, Ninglun Gu, Jin Yang, Kailai Zhang, Yelun Bao, Jun Wang·
arXiv:2508.09883v2 Announce Type: replace-cross Abstract: Large language models (LLMs) demonstrate remarkable reasoning capabilities in tasks such as algorithmic coding and mathematical problem-solving. Recent methods have improved reasoning through expanded corpus and multistage…
Retrieval-Augmented Generation (RAG) enhanced by Knowledge Graphs has shown promise in complex multi-hop reasoning tasks. However, existing graph-based retrieval methods typically rely on flat, undirected topologies. During the retrieval process, the probability flow often gets t…
