CLEANER: Self-Purified Trajectories Boost Agentic Reinforcement Learning

Agentic Reinforcement Learning (RL) has empowered Large Language Models (LLMs) to utilize tools like Python interpreters for complex problem-solving. However, for parameter-constrained models (e.g., 4B--7B), the exploration phase is often plagued by frequent execution failures, creating noisy trajectories that hinder policy optimization. Under standard outcome-based reward settings, this noise leads to a critical credit assignment issue, where erroneous actions are inadvertently reinforced alongside successful outcomes. Existing mitigations face a dilemma: dense rewards often trigger reward hacking, while supersampling incurs prohibitive computational costs. To address these challenges, we propose CLEANER. Distinct from external filtering methods, CLEANER exploits the model's intrinsic self-correction capabilities to eliminate error-contaminated context directly during data collection. At its core, the Similarity-Aware Adaptive Rollback (SAAR) mechanism autonomously constructs clean, purified trajectories by retrospectively replacing failures with successful self-corrections. Based on semantic similarity, SAAR adaptively regulates replacement granularity from shallow execution repairs to deep reasoning substitutions. By training on these self-purified paths, the model internalizes correct reasoning patterns rather than error-recovery loops. Empirical results on AIME24/25, GPQA, and LiveCodeBench show average accuracy gains of 6%, 3%, and 5% over baselines. Notably, CLEANER matches state-of-the-art performance using only one-third of the training steps, highlighting trajectory purification as a scalable solution for efficient agentic RL. Our models and code are available at GitHub

What Gets Activated: Uncovering Domain and Driver Experts in MoE Language Models

Most interpretability work focuses on layer- or neuron-level mechanisms in Transformers, leaving expert-level behavior in MoE LLMs underexplored. Motivated by functional specialization in the human brain, we analyze expert activation by distinguishing domain and driver experts. In this work, we study expert activation in MoE models across three public domains and address two key questions: (1) which experts are activated, and whether certain expert types exhibit consistent activation patterns; and (2) how tokens are associated with and trigger the activation of specific experts. To answer these questions, we introduce entropy-based and causal-effect metrics to assess whether an expert is strongly favored for a particular domain, and how strongly expert activation contributes causally to the model's output, thus identify domain and driver experts, respectively. Furthermore, we explore how individual tokens are associated with the activation of specific experts. Our analysis reveals that (1) Among the activated experts, some show clear domain preferences, while others exert strong causal influence on model performance, underscoring their decisive roles. (2) tokens occurring earlier in a sentence are more likely to trigger the driver experts, and (3) adjusting the weights of domain and driver experts leads to significant performance gains across all three models and domains. These findings shed light on the internal mechanisms of MoE models and enhance their interpretability.

Beyond Physical Labels: Redefining Domains for Robust WiFi-based Gesture Recognition

In this paper, we propose GesFi, a novel WiFi-based gesture recognition system that introduces WiFi latent domain mining to redefine domains directly from the data itself. GesFi first processes raw sensing data collected from WiFi receivers using CSI-ratio denoising, Short-Time Fast Fourier Transform, and visualization techniques to generate standardized input representations. It then employs class-wise adversarial learning to suppress gesture semantic and leverages unsupervised clustering to automatically uncover latent domain factors responsible for distributional shifts. These latent domains are then aligned through adversarial learning to support robust cross-domain generalization. Finally, the system is applied to the target environment for robust gesture inference. We deployed GesFi under both single-pair and multi-pair settings using commodity WiFi transceivers, and evaluated it across multiple public datasets and real-world environments. Compared to state-of-the-art baselines, GesFi achieves up to 78% and 50% performance improvements over existing adversarial methods, and consistently outperforms prior generalization approaches across most cross-domain tasks.

RoboMIND 2.0: A Multimodal, Bimanual Mobile Manipulation Dataset for Generalizable Embodied Intelligence

While data-driven imitation learning has revolutionized robotic manipulation, current approaches remain constrained by the scarcity of large-scale, diverse real-world demonstrations. Consequently, the ability of existing models to generalize across long-horizon bimanual tasks and mobile manipulation in unstructured environments remains limited. To bridge this gap, we present RoboMIND 2.0, a comprehensive real-world dataset comprising over 310K dual-arm manipulation trajectories collected across six distinct robot embodiments and 739 complex tasks. Crucially, to support research in contact-rich and spatially extended tasks, the dataset incorporates 12K tactile-enhanced episodes and 20K mobile manipulation trajectories. Complementing this physical data, we construct high-fidelity digital twins of our real-world environments, releasing an additional 20K-trajectory simulated dataset to facilitate robust sim-to-real transfer. To fully exploit the potential of RoboMIND 2.0, we propose MIND-2 system, a hierarchical dual-system frame-work optimized via offline reinforcement learning. MIND-2 integrates a high-level semantic planner (MIND-2-VLM) to decompose abstract natural language instructions into grounded subgoals, coupled with a low-level Vision-Language-Action executor (MIND-2-VLA), which generates precise, proprioception-aware motor actions.

ATLAS: A High-Difficulty, Multidisciplinary Benchmark for Frontier Scientific Reasoning

The rapid advancement of Large Language Models (LLMs) has led to performance saturation on many established benchmarks, questioning their ability to distinguish frontier models. Concurrently, existing high-difficulty benchmarks often suffer from narrow disciplinary focus, oversimplified answer formats, and vulnerability to data contamination, creating a fidelity gap with real-world scientific inquiry. To address these challenges, we introduce ATLAS (AGI-Oriented Testbed for Logical Application in Science), a large-scale, high-difficulty, and cross-disciplinary evaluation suite composed of approximately 800 original problems. Developed by domain experts (PhD-level and above), ATLAS spans seven core scientific fields: mathematics, physics, chemistry, biology, computer science, earth science, and materials science. Its key features include: (1) High Originality and Contamination Resistance, with all questions newly created or substantially adapted to prevent test data leakage; (2) Cross-Disciplinary Focus, designed to assess models' ability to integrate knowledge and reason across scientific domains; (3) High-Fidelity Answers, prioritizing complex, open-ended answers involving multi-step reasoning and LaTeX-formatted expressions over simple multiple-choice questions; and (4) Rigorous Quality Control, employing a multi-stage process of expert peer review and adversarial testing to ensure question difficulty, scientific value, and correctness. We also propose a robust evaluation paradigm using a panel of LLM judges for automated, nuanced assessment of complex answers. Preliminary results on leading models demonstrate ATLAS's effectiveness in differentiating their advanced scientific reasoning capabilities. We plan to develop ATLAS into a long-term, open, community-driven platform to provide a reliable "ruler" for progress toward Artificial General Intelligence.

LLM-YOLOMS: Large Language Model-based Semantic Interpretation and Fault Diagnosis for Wind Turbine Components

The health condition of wind turbine (WT) components is crucial for ensuring stable and reliable operation. However, existing fault detection methods are largely limited to visual recognition, producing structured outputs that lack semantic interpretability and fail to support maintenance decision-making. To address these limitations, this study proposes an integrated framework that combines YOLOMS with a large language model (LLM) for intelligent fault analysis and diagnosis. Specifically, YOLOMS employs multi-scale detection and sliding-window cropping to enhance fault feature extraction, while a lightweight key-value (KV) mapping module bridges the gap between visual outputs and textual inputs. This module converts YOLOMS detection results into structured textual representations enriched with both qualitative and quantitative attributes. A domain-tuned LLM then performs semantic reasoning to generate interpretable fault analyses and maintenance recommendations. Experiments on real-world datasets demonstrate that the proposed framework achieves a fault detection accuracy of 90.6\% and generates maintenance reports with an average accuracy of 89\%, thereby improving the interpretability of diagnostic results and providing practical decision support for the operation and maintenance of wind turbines.

LIFT: Interpretable truck driving risk prediction with literature-informed fine-tuned LLMs

This study proposes an interpretable prediction framework with literature-informed fine-tuned (LIFT) LLMs for truck driving risk prediction. The framework integrates an LLM-driven Inference Core that predicts and explains truck driving risk, a Literature Processing Pipeline that filters and summarizes domain-specific literature into a literature knowledge base, and a Result Evaluator that evaluates the prediction performance as well as the interpretability of the LIFT LLM. After fine-tuning on a real-world truck driving risk dataset, the LIFT LLM achieved accurate risk prediction, outperforming benchmark models by 26.7% in recall and 10.1% in F1-score. Furthermore, guided by the literature knowledge base automatically constructed from 299 domain papers, the LIFT LLM produced variable importance ranking consistent with that derived from the benchmark model, while demonstrating robustness in interpretation results to various data sampling conditions. The LIFT LLM also identified potential risky scenarios by detecting key combination of variables in truck driving risk, which were verified by PERMANOVA tests. Finally, we demonstrated the contribution of the literature knowledge base and the fine-tuning process in the interpretability of the LIFT LLM, and discussed the potential of the LIFT LLM in data-driven knowledge discovery.

Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model

We present Ring-1T, the first open-source, state-of-the-art thinking model with a trillion-scale parameter. It features 1 trillion total parameters and activates approximately 50 billion per token. Training such models at a trillion-parameter scale introduces unprecedented challenges, including train-inference misalignment, inefficiencies in rollout processing, and bottlenecks in the RL system. To address these, we pioneer three interconnected innovations: (1) IcePop stabilizes RL training via token-level discrepancy masking and clipping, resolving instability from training-inference mismatches; (2) C3PO++ improves resource utilization for long rollouts under a token budget by dynamically partitioning them, thereby obtaining high time efficiency; and (3) ASystem, a high-performance RL framework designed to overcome the systemic bottlenecks that impede trillion-parameter model training. Ring-1T delivers breakthrough results across critical benchmarks: 93.4 on AIME-2025, 86.72 on HMMT-2025, 2088 on CodeForces, and 55.94 on ARC-AGI-v1. Notably, it attains a silver medal-level result on the IMO-2025, underscoring its exceptional reasoning capabilities. By releasing the complete 1T parameter MoE model to the community, we provide the research community with direct access to cutting-edge reasoning capabilities. This contribution marks a significant milestone in democratizing large-scale reasoning intelligence and establishes a new baseline for open-source model performance.

Enabling MoE on the Edge via Importance-Driven Expert Scheduling

The Mixture of Experts (MoE) architecture has emerged as a key technique for scaling Large Language Models by activating only a subset of experts per query. Deploying MoE on consumer-grade edge hardware, however, is constrained by limited device memory, making dynamic expert offloading essential. Unlike prior work that treats offloading purely as a scheduling problem, we leverage expert importance to guide decisions, substituting low-importance activated experts with functionally similar ones already cached in GPU memory, thereby preserving accuracy. As a result, this design reduces memory usage and data transfer, while largely eliminating PCIe overhead. In addition, we introduce a scheduling policy that maximizes the reuse ratio of GPU-cached experts, further boosting efficiency. Extensive evaluations show that our approach delivers 48% lower decoding latency with over 60% expert cache hit rate, while maintaining nearly lossless accuracy.

CRABS: A syntactic-semantic pincer strategy for bounding LLM interpretation of Python notebooks

Recognizing the information flows and operations comprising data science and machine learning Python notebooks is critical for evaluating, reusing, and adapting notebooks for new tasks. Investigating a notebook via re-execution often is impractical due to the challenges of resolving data and software dependencies. While Large Language Models (LLMs) pre-trained on large codebases have demonstrated effectiveness in understanding code without running it, we observe that they fail to understand some realistic notebooks due to hallucinations and long-context challenges. To address these issues, we propose a notebook understanding task yielding an information flow graph and corresponding cell execution dependency graph for a notebook, and demonstrate the effectiveness of a pincer strategy that uses limited syntactic analysis to assist full comprehension of the notebook using an LLM. Our Capture and Resolve Assisted Bounding Strategy (CRABS) employs shallow syntactic parsing and analysis of the abstract syntax tree (AST) to capture the correct interpretation of a notebook between lower and upper estimates of the inter-cell I/O sets, then uses an LLM to resolve remaining ambiguities via cell-by-cell zero-shot learning, thereby identifying the true data inputs and outputs of each cell. We evaluate and demonstrate the effectiveness of our approach using an annotated dataset of 50 representative, highly up-voted Kaggle notebooks that together represent 3454 actual cell inputs and outputs. The LLM correctly resolves 1397 of 1425 (98%) ambiguities left by analyzing the syntactic structure of these notebooks. Across 50 notebooks, CRABS achieves average F1 scores of 98% identifying cell-to-cell information flows and 99% identifying transitive cell execution dependencies.