相关论文: Guided Self-Evolving LLMs with Minimal Human Super…

Large language models (LLMs) have improved significantly in their reasoning through extensive training on massive datasets. However, relying solely on additional data for improvement is becoming increasingly impractical, highlighting the…

Training a task-specific small reasoning model is challenging when direct human supervision or high-quality labels are scarce. However, LLMs with reasoning capabilities produce abundant intermediate reasoning traces that can be…

Creating high-quality data for training robust language-instructed agents is a long-lasting challenge in embodied AI. In this paper, we introduce a Self-Refining Data Flywheel (SRDF) that generates high-quality and large-scale navigational…

Large Language Models (LLMs) have demonstrated remarkable progress in complex reasoning tasks through both post-training and test-time scaling laws. While prevalent test-time scaling approaches are often realized by using external reward…

While language models have shown remarkable performance across diverse tasks, they still encounter challenges in complex reasoning scenarios. Recent research suggests that language models trained on linearized search traces toward…

Scaling up self-supervised learning has driven breakthroughs in language and vision, yet comparable progress has remained elusive in reinforcement learning (RL). In this paper, we study building blocks for self-supervised RL that unlock…

We present a general framework for evolutionary learning to emergent unbiased state representation without any supervision. Evolutionary frameworks such as self-play converge to bad local optima in case of multi-agent reinforcement learning…

Prompt-based continual learning provides a rehearsal-free solution by tuning small sets of parameters while keeping pre-trained models frozen. To meet the complex demands of sequential tasks, it is crucial to integrate task-specific…

Traditional methods for automating recommender system design, such as Neural Architecture Search (NAS), are often constrained by a fixed search space defined by human priors, limiting innovation to pre-defined operators. While recent…

Large language models often struggle with length generalization and solving complex problem instances beyond their training distribution. We present a self-improvement approach where models iteratively generate and learn from their own…

Recently, self-supervised learning methods like MoCo, SimCLR, BYOL and SwAV have reduced the gap with supervised methods. These results have been achieved in a control environment, that is the highly curated ImageNet dataset. However, the…

Drifting, characterized by controlled vehicle motion at high sideslip angles, is crucial for safely handling emergency scenarios at the friction limits. While recent reinforcement learning approaches show promise for drifting control, they…

Significant progress in reward modeling over recent years has been driven by a paradigm shift from task-specific designs towards generalist reward models. Despite this trend, developing effective reward models remains a fundamental…

Reflection is widely recognized as a cornerstone of student development, fostering critical thinking, self-regulation, and deep conceptual understanding. Traditionally, reflective skills have been cultivated through structured feedback,…

Co-evolutionary self-play, where one language model generates problems and another solves them, promises autonomous curriculum learning without human supervision. In practice, the proposer quickly converges to a narrow distribution of…

Existing low-light image enhancement (LIE) methods have achieved noteworthy success in solving synthetic distortions, yet they often fall short in practical applications. The limitations arise from two inherent challenges in real-world LIE:…

Recent advances in test-time scaling have led to the emergence of thinking LLMs that exhibit self-reflective behaviors and multi-step reasoning. While RL drives this self-improvement paradigm, a recent study (Gandhi et al., 2025) shows that…

Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of LLMs, yet a fundamental limitation remains: models cannot learn from problems that are too difficult to solve under their current policy, as these…

Reward modeling (RM), which captures human preferences to align large language models (LLMs), is increasingly employed in tasks such as model finetuning, response filtering, and ranking. However, due to the inherent complexity of human…

When asked a question in a language less seen in its training data, current reasoning large language models (RLMs) often exhibit dramatically lower performance than when asked the same question in English. In response, we introduce…