Ring-lite: Scalable Reasoning via C3PO-Stabilized Reinforcement Learning for LLMs

We present Ring-lite, a Mixture-of-Experts (MoE)-based large language model optimized via reinforcement learning (RL) to achieve efficient and robust reasoning capabilities. Built upon the publicly available Ling-lite model, a 16.8 billion parameter model with 2.75 billion activated parameters, our approach matches the performance of state-of-the-art (SOTA) small-scale reasoning models on challenging benchmarks (e.g., AIME, LiveCodeBench, GPQA-Diamond) while activating only one-third of the parameters required by comparable models. To accomplish this, we introduce a joint training pipeline integrating distillation with RL, revealing undocumented challenges in MoE RL training. First, we identify optimization instability during RL training, and we propose Constrained Contextual Computation Policy Optimization(C3PO), a novel approach that enhances training stability and improves computational throughput via algorithm-system co-design methodology. Second, we empirically demonstrate that selecting distillation checkpoints based on entropy loss for RL training, rather than validation metrics, yields superior performance-efficiency trade-offs in subsequent RL training. Finally, we develop a two-stage training paradigm to harmonize multi-domain data integration, addressing domain conflicts that arise in training with mixed dataset. We will release the model, dataset, and code.

Distributed Activity Detection for Cell-Free Hybrid Near-Far Field Communications

A great amount of endeavor has recently been devoted to activity detection for massive machine-type communications in cell-free massive MIMO. However, in practice, as the number of antennas at the access points (APs) increases, the Rayleigh distance that separates the near-field and far-field regions also expands, rendering the conventional assumption of far-field propagation alone impractical. To address this challenge, this paper considers a hybrid near-far field activity detection in cell-free massive MIMO, and establishes a covariance-based formulation, which facilitates the development of a distributed algorithm to alleviate the computational burden at the central processing unit (CPU). Specifically, each AP performs local activity detection for the devices and then transmits the detection result to the CPU for further processing. In particular, a novel coordinate descent algorithm based on the Sherman-Morrison-Woodbury update with Taylor expansion is proposed to handle the local detection problem at each AP. Moreover, we theoretically analyze how the hybrid near-far field channels affect the detection performance. Simulation results validate the theoretical analysis and demonstrate the superior performance of the proposed approach compared with existing approaches.

VLM-SFD: VLM-Assisted Siamese Flow Diffusion Framework for Dual-Arm Cooperative Manipulation

Dual-arm cooperative manipulation holds great promise for tackling complex real-world tasks that demand seamless coordination and adaptive dynamics. Despite substantial progress in learning-based motion planning, most approaches struggle to generalize across diverse manipulation tasks and adapt to dynamic, unstructured environments, particularly in scenarios involving interactions between two objects such as assembly, tool use, and bimanual grasping. To address these challenges, we introduce a novel VLM-Assisted Siamese Flow Diffusion (VLM-SFD) framework for efficient imitation learning in dual-arm cooperative manipulation. The proposed VLM-SFD framework exhibits outstanding adaptability, significantly enhancing the ability to rapidly adapt and generalize to diverse real-world tasks from only a minimal number of human demonstrations. Specifically, we propose a Siamese Flow Diffusion Network (SFDNet) employs a dual-encoder-decoder Siamese architecture to embed two target objects into a shared latent space, while a diffusion-based conditioning process-conditioned by task instructions-generates two-stream object-centric motion flows that guide dual-arm coordination. We further design a dynamic task assignment strategy that seamlessly maps the predicted 2D motion flows into 3D space and incorporates a pre-trained vision-language model (VLM) to adaptively assign the optimal motion to each robotic arm over time. Experiments validate the effectiveness of the proposed method, demonstrating its ability to generalize to diverse manipulation tasks while maintaining high efficiency and adaptability. The code and demo videos are publicly available on our project website https://sites.google.com/view/vlm-sfd/.

MMWiLoc: A Multi-Sensor Dataset and Robust Device-Free Localization Method Using Commercial Off-The-Shelf Millimeter Wave Wi-Fi Devices

Device-free Wi-Fi sensing has numerous benefits in practical settings, as it eliminates the requirement for dedicated sensing devices and can be accomplished using current low-cost Wi-Fi devices. With the development of Wi-Fi standards, millimeter wave Wi-Fi devices with 60GHz operating frequency and up to 4GHz bandwidth have become commercially available. Although millimeter wave Wi-Fi presents great promise for Device-Free Wi-Fi sensing with increased bandwidth and beam-forming ability, there still lacks a method for localization using millimeter wave Wi-Fi. Here, we present two major contributions: First, we provide a comprehensive multi-sensor dataset that synchronously captures human movement data from millimeter wave Wi-Fi, 2.4GHz Wi-Fi, and millimeter wave radar sensors. This dataset enables direct performance comparisons across different sensing modalities and facilitates reproducible researches in indoor localization. Second, we introduce MMWiLoc, a novel localization method that achieves centimeter-level precision with low computational cost. MMWiLoc incorporates two components: beam pattern calibration using Expectation Maximization and target localization through Multi-Scale Compression Sensing. The system processes beam Signal-to-Noise Ratio (beamSNR) information from the beam-forming process to determine target Angle of Arrival (AoA), which is then fused across devices for localization. Our extensive evaluation demonstrates that MMWiLoc achieves centimeter-level precision, outperforming 2.4GHz Wi-Fi systems while maintaining competitive performance with high-precision radar systems. The dataset and examples processing code will be released after this paper is accepted at https://github.com/wowoyoho/MMWiLoc.

Shapley Machine: A Game-Theoretic Framework for N-Agent Ad Hoc Teamwork

Open multi-agent systems are increasingly important in modeling real-world applications, such as smart grids, swarm robotics, etc. In this paper, we aim to investigate a recently proposed problem for open multi-agent systems, referred to as n-agent ad hoc teamwork (NAHT), where only a number of agents are controlled. Existing methods tend to be based on heuristic design and consequently lack theoretical rigor and ambiguous credit assignment among agents. To address these limitations, we model and solve NAHT through the lens of cooperative game theory. More specifically, we first model an open multi-agent system, characterized by its value, as an instance situated in a space of cooperative games, generated by a set of basis games. We then extend this space, along with the state space, to accommodate dynamic scenarios, thereby characterizing NAHT. Exploiting the justifiable assumption that basis game values correspond to a sequence of n-step returns with different horizons, we represent the state values for NAHT in a form similar to $\lambda$-returns. Furthermore, we derive Shapley values to allocate state values to the controlled agents, as credits for their contributions to the ad hoc team. Different from the conventional approach to shaping Shapley values in an explicit form, we shape Shapley values by fulfilling the three axioms uniquely describing them, well defined on the extended game space describing NAHT. To estimate Shapley values in dynamic scenarios, we propose a TD($\lambda$)-like algorithm. The resulting reinforcement learning (RL) algorithm is referred to as Shapley Machine. To our best knowledge, this is the first time that the concepts from cooperative game theory are directly related to RL concepts. In experiments, we demonstrate the effectiveness of Shapley Machine and verify reasonableness of our theory.

From Judgment to Interference: Early Stopping LLM Harmful Outputs via Streaming Content Monitoring

Though safety alignment has been applied to most large language models (LLMs), LLM service providers generally deploy a subsequent moderation as the external safety guardrail in real-world products. Existing moderators mainly practice a conventional full detection, which determines the harmfulness based on the complete LLM output, causing high service latency. Recent works pay more attention to partial detection where moderators oversee the generation midway and early stop the output if harmfulness is detected, but they directly apply moderators trained with the full detection paradigm to incomplete outputs, introducing a training-inference gap that lowers the performance. In this paper, we explore how to form a data-and-model solution that natively supports partial detection. For the data, we construct FineHarm, a dataset consisting of 29K prompt-response pairs with fine-grained annotations to provide reasonable supervision for token-level training. Then, we propose the streaming content monitor, which is trained with dual supervision of response- and token-level labels and can follow the output stream of LLM to make a timely judgment of harmfulness. Experiments show that SCM gains 0.95+ in macro F1 score that is comparable to full detection, by only seeing the first 18% of tokens in responses on average. Moreover, the SCM can serve as a pseudo-harmfulness annotator for improving safety alignment and lead to a higher harmlessness score than DPO.

Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation

We present Chain-of-Action (CoA), a novel visuo-motor policy paradigm built upon Trajectory Autoregressive Modeling. Unlike conventional approaches that predict next step action(s) forward, CoA generates an entire trajectory by explicit backward reasoning with task-specific goals through an action-level Chain-of-Thought (CoT) process. This process is unified within a single autoregressive structure: (1) the first token corresponds to a stable keyframe action that encodes the task-specific goals; and (2) subsequent action tokens are generated autoregressively, conditioned on the initial keyframe and previously predicted actions. This backward action reasoning enforces a global-to-local structure, allowing each local action to be tightly constrained by the final goal. To further realize the action reasoning structure, CoA incorporates four complementary designs: continuous action token representation; dynamic stopping for variable-length trajectory generation; reverse temporal ensemble; and multi-token prediction to balance action chunk modeling with global structure. As a result, CoA gives strong spatial generalization capabilities while preserving the flexibility and simplicity of a visuo-motor policy. Empirically, we observe CoA achieves the state-of-the-art performance across 60 RLBench tasks and 8 real-world manipulation tasks.

Automating Exploratory Multiomics Research via Language Models

This paper introduces PROTEUS, a fully automated system that produces data-driven hypotheses from raw data files. We apply PROTEUS to clinical proteogenomics, a field where effective downstream data analysis and hypothesis proposal is crucial for producing novel discoveries. PROTEUS uses separate modules to simulate different stages of the scientific process, from open-ended data exploration to specific statistical analysis and hypothesis proposal. It formulates research directions, tools, and results in terms of relationships between biological entities, using unified graph structures to manage complex research processes. We applied PROTEUS to 10 clinical multiomics datasets from published research, arriving at 360 total hypotheses. Results were evaluated through external data validation and automatic open-ended scoring. Through exploratory and iterative research, the system can navigate high-throughput and heterogeneous multiomics data to arrive at hypotheses that balance reliability and novelty. In addition to accelerating multiomic analysis, PROTEUS represents a path towards tailoring general autonomous systems to specialized scientific domains to achieve open-ended hypothesis generation from data.

FlowOE: Imitation Learning with Flow Policy from Ensemble RL Experts for Optimal Execution under Heston Volatility and Concave Market Impacts

Optimal execution in financial markets refers to the process of strategically transacting a large volume of assets over a period to achieve the best possible outcome by balancing the trade-off between market impact costs and timing or volatility risks. Traditional optimal execution strategies, such as static Almgren-Chriss models, often prove suboptimal in dynamic financial markets. This paper propose flowOE, a novel imitation learning framework based on flow matching models, to address these limitations. FlowOE learns from a diverse set of expert traditional strategies and adaptively selects the most suitable expert behavior for prevailing market conditions. A key innovation is the incorporation of a refining loss function during the imitation process, enabling flowOE not only to mimic but also to improve upon the learned expert actions. To the best of our knowledge, this work is the first to apply flow matching models in a stochastic optimal execution problem. Empirical evaluations across various market conditions demonstrate that flowOE significantly outperforms both the specifically calibrated expert models and other traditional benchmarks, achieving higher profits with reduced risk. These results underscore the practical applicability and potential of flowOE to enhance adaptive optimal execution.

Recycling the Web: A Method to Enhance Pre-training Data Quality and Quantity for Language Models

Scaling laws predict that the performance of large language models improves with increasing model size and data size. In practice, pre-training has been relying on massive web crawls, using almost all data sources publicly available on the internet so far. However, this pool of natural data does not grow at the same rate as the compute supply. Furthermore, the availability of high-quality texts is even more limited: data filtering pipelines often remove up to 99% of the initial web scrapes to achieve state-of-the-art. To address the "data wall" of pre-training scaling, our work explores ways to transform and recycle data discarded in existing filtering processes. We propose REWIRE, REcycling the Web with guIded REwrite, a method to enrich low-quality documents so that they could become useful for training. This in turn allows us to increase the representation of synthetic data in the final pre-training set. Experiments at 1B, 3B and 7B scales of the DCLM benchmark show that mixing high-quality raw texts and our rewritten texts lead to 1.0, 1.3 and 2.5 percentage points improvement respectively across 22 diverse tasks, compared to training on only filtered web data. Training on the raw-synthetic data mix is also more effective than having access to 2x web data. Through further analysis, we demonstrate that about 82% of the mixed in texts come from transforming lower-quality documents that would otherwise be discarded. REWIRE also outperforms related approaches of generating synthetic data, including Wikipedia-style paraphrasing, question-answer synthesizing and knowledge extraction. These results suggest that recycling web texts holds the potential for being a simple and effective approach for scaling pre-training data.