Reflective Recovery is a self‑supervised method that turns failed reasoning attempts into training data, enabling large language models to learn how to correct mistakes during inference. By extracting initial segments of erroneous trajectories and using them as prompts, the approach teaches models to recognize and recover from errors without external critics. Experiments show significant accuracy gains on benchmarks such as AIME 2025 and Minerva, and the method overcomes the scaling collapse problem, fostering emergent self‑correction behaviors.
By Qirui Chen, Renjie Pi, Jiahui Gao, Lingpeng Kong
arXiv:2607. 07492v1 Announce Type: new Abstract: Many reasoning tasks are not well described by a single left-to-right chain: a solver may need to pursue a plausible branch, observe delayed failure, and return to the latest prefix that can still be completed.
By Dmitry Beresnev, Vladimir Makharev, Roman Khalikov, Ivan Oseledets, Petr Anokhin
The paper introduces Experience Memory Graph (EMG), a framework that transforms agent failure recovery into a graph matching problem. By converting failed and successful trajectories into directed action decision graphs, EMG extracts common subgraphs and edit paths that reveal explicit correction strategies, storing them in a memory graph with intra‑task nodes and cross‑task edges. During testing, EMG retrieves relevant insights to guide the agent in a single attempt, improving performance on ALFWorld and ScienceWorld without extra reflection rounds.
By Wenjun Wang, Yuchen Fang, Fengrui Liu, Zibo Liang, Kai Zheng
arXiv:2610.19561v1 Announce Type: new
Abstract: An early mistake in the generated reasoning can constrain all subsequent reasoning. Repeated-generation methods, like MCMC-based approaches, attempt to...
By Yi Ding, Mengyu Wang, Min Zhou, Menglin Yang
arXiv:2606. 17524v1 Announce Type: new Abstract: Large language models show strong reasoning ability, but their internal reasoning process can remain unstable in complex multi-step settings, where early hidden-state errors may propagate to incorrect predictions.
By Chia-Hsuan Hsu, Jui-Ming Yao
arXiv:2607. 01595v1 Announce Type: new Abstract: As the scale and complexity of cloud-based AI systems continue to escalate, ensuring service reliability through rapid fault detection and adaptive recovery has become a critical challenge.
By Junyan Tan, Haoran Lin, Siyuan Guo, Yichen Fang, Xinyue Luo, Tianyu Shen, Zeyu Qiao
ARBOR is a framework that tracks reasoning states as a tree throughout a model’s chain of thought, enabling attribution of failures such as overthinking, underthinking, and unsupported deductions to specific state-step relations. By maintaining a global view of the state trajectory, ARBOR distinguishes failure steps from at‑risk steps where the failure did not occur, refining failure modes into distinct subtypes. The method can be integrated with existing inference-time mitigations and training data, and its code, data, and interactive examples are publicly available.
By Huihan Li, Siyuan Wang, Naga Vamsi Ramana Dinavahi, Xiaohui Liu, Xiang Ren
arXiv:2607. 13884v1 Announce Type: new Abstract: Large Language Model (LLM) agents have shown remarkable capabilities in autonomous decision-making by generating sequential trajectories of states, actions, and observations.
By Wenjun Wang, Yuchen Fang, Fengrui Liu, Zibo Liang, Kai Zheng
arXiv:2602. 02416v2 Announce Type: replace Abstract: Self-correction in language models remains elusive.
By Ankur Samanta, Akshayaa Magesh, Ayush Jain, Kavosh Asadi, Youliang Yu, Daniel Jiang, Boris Vidolov, Kaveh Hassani, Paul Sajda, Jalaj Bhandari, Yonathan Efroni
An early mistake in the generated reasoning can constrain all subsequent reasoning. Repeated-generation methods, like MCMC-based approaches, attempt to improve output quality by sampling multiple gene...
The paper introduces epistemic memory, a validity-maintenance layer for intelligent systems that tracks when stored knowledge remains applicable. It formalizes a dynamic epistemic quotient and shows that fixed semantic representations inevitably incur error as epistemic boundaries shift. The authors propose Observable Belief Memory (OBM), which combines current epistemic quotients, belief over quotient classes, and within-class provenance, and demonstrate that explicit epistemic tracking improves robustness under changing observation conditions.
By Pin-Han Ho, Limei Peng, Yiming Miao, Yan Jiao
arXiv:2607. 17188v2 Announce Type: replace Abstract: While test-time scaling improves the problem-solving ability of large reasoning models (LRMs) through additional inference-time computation, it can also exacerbate overthinking and underthinking, which we formulate as reasoning state--action mismatch.
By Cheng Yan, Zhijun Fan, Guangyang Ye, Fan Xu, Xiang Xia, Yawei Wang, Wuyang Zhang