arXiv subjects

Hailin Zhang

Publications and source records attributed to Hailin Zhang.

At least 19 recordsLinked to original sources

arXiv2026

An Ultra-Low Profile Metasurface for Enhanced Non-Diffractive OAM Beam Generation Based on Sparse Feed Array

This paper proposes a novel ultra-low profile transmitted metasurface to generate enhanced non-diffractive orbital angular momentum (OAM) beams, employing a sparse feed array (SFA) to create a quasi-plane wave excitation for the first time. Our simulation indicates that with uniform amplitude excitation, the non-diffraction performance of Bessel beam produced by metasurface, surpasses that of conventional single-feed excitation. Based on this principle, a 5 * 5 sparse feed array is introduced and positioned less than one wavelength from the metasurface, ensuring a quasi-uniform amplitude excitation across all units. Further, this metasurface leverages its flexible phase control capability and integrates the spatial phase, OAM phase, and axicon phase to generate an enhanced high-order Bessel beam. The simulated results confirm a successful non-diffractive Bessel beam generation carrying OAM with mode l = +2, exhibiting reduced beam divergence and higher gain. This design also offers benefits of ultra-low profile, high aperture efficiency, low structural complexity.

eess.SP↗

arXiv2026

A Wideband Multibeam Planar Quasi-Yagi Array Antenna Based on Parasitic Pixel Strips

This letter presents a low-complexity parasitic pixel strip (PPS) structure for array antenna to generate six radiations with flexible beam control. To realize multibeam operation, three groups of symmetrical strip-shaped pixels are integrated with the feed antenna, the connection state of which can be controlled by PIN diode. Leveraging the reconfigurable parasitic configuration, the coupling path of the antenna can be strategically manipulated, achieving the desired multibeam property. Thus, by utilizing the PPS structure in conjunction with a planar quasi-Yagi antenna, this single antenna can operate at four states with beam direction towards 40{\deg}, -40{\deg}, 0{\deg}, along with a dual-beam at +/-45{\deg}, within the frequency range from 5.0 to 5.6 GHz. Moreover, a prototype of a 1 * 2 array with the PPS structure is also fabricated to achieve six beams with a scan range of 40{\deg}, 20{\deg}, 0{\deg}, -20{\deg}, -40{\deg} and a dual-beam at +/-30{\deg}. Notably, only four PIN diodes are utilized to implement six radiations, verifying that this approach effectively minimizes the excessive use of switches and extends the variety of beams, which provides a low complexity method for multibeam array antenna without complicated beamforming network. These antennas satisfy the benefits of wideband, planner structure, low structural complexity, low cost, and flexible beam control.

eess.SP↗

arXiv2026

Full-Pipeline Inference Optimization for MiMo-V2.5 Series: Pushing Hybrid SWA Efficiency to the Limit

We present a full-pipeline inference optimization for the MiMo-V2.5 model family, which combines Hybrid Sliding Window Attention (Hybrid SWA), sparse Mixture-of-Experts (MoE), and multimodal encoders. While Hybrid SWA can ideally reduce both attention compute and KVCache storage significantly compared to Full Attention, realizing these gains in production requires substantial engineering effort. We systematically optimize the KVCache system with layerwise prefetch, SWA-aware prefix cache trees, and specialized placement strategies, achieving strict $O(W)$ SWA storage and high cache hit rates. We further build GCache, a high-performance distributed cache infrastructure with RDMA-optimized networking, and develop a KVCache-affinity router to reduce computation while preserving load balancing. We also optimize for multimodal inputs, including GPU image preprocessing, parallel video decoding, and multimodal cache sharing. Together, these optimizations constitute the first large-scale LLM serving system in production that efficiently covers the Hybrid SWA + MoE + multimodal composite architecture.

cs.AR↗

arXiv2026

RaBitQCache: Rotated Binary Quantization for KVCache in Long Context LLM Inference

Long-context Large Language Model inference is severely bottlenecked by the massive Key-Value (KV) cache, yet existing sparse attention methods often suffer from static fixed-budget (Top-k) retrieval or rely on proxy scores that are computationally expensive and biased. To address these limitations, we propose RaBitQCache, a novel sparse attention framework that utilizes randomized rotated binary quantization and high-throughput binary-INT4 arithmetic to efficiently estimate attention weights. Our proxy score serves as an unbiased estimator with a proven error bound, enabling adaptive Top-p retrieval that dynamically adjusts the token budget based on actual attention sparsity. We further implement a hardware-aware system with asynchronous pipelining and lazy updates to mask overhead. Evaluations demonstrate that RaBitQCache significantly accelerates inference and reduces memory I/O while preserving generation quality compared to state-of-the-art baselines. Code is available at https://github.com/Sakuraaa0/RaBitQCache.git.

cs.LG↗

arXiv2026

MOPD: Multi-Teacher On-Policy Distillation for Capability Integration in LLM Post-Training

Modern large language models (LLMs) rely on reinforcement learning during post-training to push specific capabilities, yet integrating multiple capabilities into one model remains hard. Existing methods, such as Off-Policy Finetune and Mix-RL, are either inefficient or lose performance. In this work, we propose Multi-teacher On-Policy Distillation (MOPD), a post-training paradigm for combining the capabilities of multiple domain RL teachers: we first run per-domain specialised RL to obtain a set of domain teachers, then distill these teachers into the student on its own rollouts. This eliminates exposure bias and provides a dense optimization signal. On Qwen3-30B-A3B, MOPD outperforms Mix-RL, Cascade RL, Off-Policy Finetune, and Param-Merge baselines, inheriting nearly all of each teacher's capability. MOPD also enables parallel, independent development of domain teachers, removing the cross-domain coupling typical of multi-domain post-training. MOPD has been deployed in the post-training of MiMo-V2-Flash, an industrial-scale frontier model, demonstrating its practical value for capability integration in frontier-scale LLMs.

cs.CL↗

arXiv2026

ARL-Tangram: Unleash the Resource Efficiency in Agentic Reinforcement Learning

Agentic reinforcement learning (RL) has emerged as a transformative workload in cloud clusters, enabling large language models (LLMs) to solve complex problems through interactions with real world. However, unlike traditional RL, agentic RL demands substantial external cloud resources, e.g., CPUs for code execution and GPUs for reward models, that exist outside the primary training cluster. Existing agentic RL framework typically rely on static over-provisioning, i.e., resources are often tied to long-lived trajectories or isolated by tasks, which leads to severe resource inefficiency. We propose the action-level orchestration, and incorporate it into ARL-Tangram, a unified resource management system that enables fine-grained external resource sharing and elasticity. ARL-Tangram utilizes a unified action-level formulation and an elastic scheduling algorithm to minimize action completion time (ACT) while satisfying heterogeneous resource constraints. Further, heterogeneous resource managers are tailored to efficiently support the action-level execution on resources with heterogeneous characteristics and topologies. Evaluation on real-world agentic RL tasks demonstrates that ARL-Tangram improves average ACT by up to 4.3$\times$, speeds up the step duration of RL training by up to 1.5$\times$, and saves the external resources by up to 71.2$\%$. This system has been deployed to support the training of the MiMo series models.

cs.DC↗

arXiv2026

EEG-Based Brain-LLM Interface for Human Preference Aligned Generation

Large language models (LLMs) are becoming an increasingly important component of human--computer interaction, enabling users to coordinate a wide range of intelligent agents through natural language. While language-based interfaces are powerful and flexible, they implicitly assume that users can reliably produce explicit linguistic input, an assumption that may not hold for users with speech or motor impairments, e.g., Amyotrophic Lateral Sclerosis (ALS). In this work, we investigate whether neural signals can be used as an alternative input to LLMs, particularly to support those socially marginalized or underserved users. We build a simple brain-LLM interface, which uses EEG signals to guide image generation models at test time. Specifically, we first train a classifier to estimate user satisfaction from EEG signals. Its predictions are then incorporated into a test-time scaling (TTS) framework that dynamically adapts model inference using neural feedback collected during user evaluation. The experiments show that EEG can predict user satisfaction, suggesting that neural activity carries information on real-time preference inference. These findings provide a first step toward integrating neural feedback into adaptive language-model inference, and hopefully open up new possibilities for future research on adaptive LLM interaction.

eess.SP↗

arXiv2026

HySparse: A Hybrid Sparse Attention Architecture with Oracle Token Selection and KV Cache Sharing

This work introduces Hybrid Sparse Attention (HySparse), a new architecture that interleaves each full attention layer with several sparse attention layers. While conceptually simple, HySparse strategically derives each sparse layer's token selection and KV caches directly from the preceding full attention layer. This architecture resolves two fundamental limitations of prior sparse attention methods. First, conventional approaches typically rely on additional proxies to predict token importance, introducing extra complexity and potentially suboptimal performance. In contrast, HySparse uses the full attention layer as a precise oracle to identify important tokens. Second, existing sparse attention designs often reduce computation without saving KV cache. HySparse enables sparse attention layers to reuse the full attention KV cache, thereby reducing both computation and memory. We evaluate HySparse on both 7B dense and 80B MoE models. Across all settings, HySparse consistently outperforms both full attention and hybrid SWA baselines. Notably, in the 80B MoE model with 49 total layers, only 5 layers employ full attention, yet HySparse achieves substantial performance gains while reducing KV cache storage by nearly 10x.

cs.CL↗

arXiv2026

Reinforcement Learning for Chain of Thought Compression with One-Domain-to-All Generalization

Chain-of-thought reasoning in large language models can trigger an "overthinking trap": longer rollouts raise cost and latency yet often yield unreliable accuracy gains. Existing methods use global, static controls that may suppress needed reasoning. We propose mastery-gated, sample-level, soft reinforcement learning compression that penalizes long rollouts only when the model already solves the problem and has produced a shorter rollout. Across benchmarks, it cuts response length by 20-40% with comparable or higher accuracy and generalizes across domains: a model trained on math spontaneously shortens unseen tasks (code, instruction following, general-knowledge QA) without hurting accuracy. We further show two-way transfer between non-agent CoT and tool-use agents: non-agent training reduces SWE-Bench Verified rounds by 13%, while compressing a thinking agent cuts SWE trajectories by 67% tokens and 52% rounds and shortens non-agent outputs by up to 44%. Compression is thus not cosmetic brevity, but an inherent computation policy -- what to keep, and what to forget.

cs.CL↗

arXiv2026

JudgeRLVR: Judge First, Generate Second for Efficient Reasoning

Reinforcement Learning with Verifiable Rewards (RLVR) has become a standard paradigm for reasoning in Large Language Models. However, optimizing solely for final-answer correctness often drives models into aimless, verbose exploration, where they rely on exhaustive trial-and-error tactics rather than structured planning to reach solutions. While heuristic constraints like length penalties can reduce verbosity, they often truncate essential reasoning steps, creating a difficult trade-off between efficiency and verification. In this paper, we argue that discriminative capability is a prerequisite for efficient generation: by learning to distinguish valid solutions, a model can internalize a guidance signal that prunes the search space. We propose JudgeRLVR, a two-stage judge-then-generate paradigm. In the first stage, we train the model to judge solution responses with verifiable answers. In the second stage, we fine-tune the same model with vanilla generating RLVR initialized from the judge. Compared to Vanilla RLVR using the same math-domain training data, JudgeRLVR achieves a better quality--efficiency trade-off for Qwen3-30B-A3B: on in-domain math, it delivers about +3.7 points average accuracy gain with -42\% average generation length; on out-of-domain benchmarks, it delivers about +4.5 points average accuracy improvement, demonstrating enhanced generalization.

cs.CL↗

arXiv2026

MiMo-V2-Flash Technical Report

We present MiMo-V2-Flash, a Mixture-of-Experts (MoE) model with 309B total parameters and 15B active parameters, designed for fast, strong reasoning and agentic capabilities. MiMo-V2-Flash adopts a hybrid attention architecture that interleaves Sliding Window Attention (SWA) with global attention, with a 128-token sliding window under a 5:1 hybrid ratio. The model is pre-trained on 27 trillion tokens with Multi-Token Prediction (MTP), employing a native 32k context length and subsequently extended to 256k. To efficiently scale post-training compute, MiMo-V2-Flash introduces a novel Multi-Teacher On-Policy Distillation (MOPD) paradigm. In this framework, domain-specialized teachers (e.g., trained via large-scale reinforcement learning) provide dense and token-level reward, enabling the student model to perfectly master teacher expertise. MiMo-V2-Flash rivals top-tier open-weight models such as DeepSeek-V3.2 and Kimi-K2, despite using only 1/2 and 1/3 of their total parameters, respectively. During inference, by repurposing MTP as a draft model for speculative decoding, MiMo-V2-Flash achieves up to 3.6 acceptance length and 2.6x decoding speedup with three MTP layers. We open-source both the model weights and the three-layer MTP weights to foster open research and community collaboration.

cs.CL↗

arXiv2025

MiMo-Audio: Audio Language Models are Few-Shot Learners

Existing audio language models typically rely on task-specific fine-tuning to accomplish particular audio tasks. In contrast, humans are able to generalize to new audio tasks with only a few examples or simple instructions. GPT-3 has shown that scaling next-token prediction pretraining enables strong generalization capabilities in text, and we believe this paradigm is equally applicable to the audio domain. By scaling MiMo-Audio's pretraining data to over one hundred million of hours, we observe the emergence of few-shot learning capabilities across a diverse set of audio tasks. We develop a systematic evaluation of these capabilities and find that MiMo-Audio-7B-Base achieves SOTA performance on both speech intelligence and audio understanding benchmarks among open-source models. Beyond standard metrics, MiMo-Audio-7B-Base generalizes to tasks absent from its training data, such as voice conversion, style transfer, and speech editing. MiMo-Audio-7B-Base also demonstrates powerful speech continuation capabilities, capable of generating highly realistic talk shows, recitations, livestreaming and debates. At the post-training stage, we curate a diverse instruction-tuning corpus and introduce thinking mechanisms into both audio understanding and generation. MiMo-Audio-7B-Instruct achieves open-source SOTA on audio understanding benchmarks (MMSU, MMAU, MMAR, MMAU-Pro), spoken dialogue benchmarks (Big Bench Audio, MultiChallenge Audio) and instruct-TTS evaluations, approaching or surpassing closed-source models. Model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-Audio.

cs.CL↗

arXiv2025

Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers

Reinforcement learning (RL) has emerged as a crucial approach for enhancing the capabilities of large language models. However, in Mixture-of-Experts (MoE) models, the routing mechanism often introduces instability, even leading to catastrophic RL training collapse. We analyze the training-inference consistency of MoE models and identify a notable discrepancy in routing behaviors between the two phases. Moreover, even under identical conditions, the routing framework can yield divergent expert selections across repeated forward passes. To address this foundational inconsistency, we propose Rollout Routing Replay (R3), a method that records routing distributions from the inference engine and replays them during training. R3 significantly reduces training-inference policy KL divergence and mitigates extreme discrepancies without compromising training speed. Extensive experiments on various settings confirm that R3 succeeds in stabilizing RL training, preventing collapse and outperforming methods such as GSPO and TIS. We believe this work can offer a new solution for stabilizing RL in MoE models.

cs.CL↗

arXiv2025

Airway Mucus Rheology: Physical Insights for Navigating through Health to Pathology and Clinical Applications

Airway mucus is a complex gel with an anisotropic three-dimensional network structure. As a crucial component of the respiratory defense barrier, it plays a vital role in maintaining airway hydration and supporting the function of airway epithelial cells. Through linear and nonlinear rheological mechanisms such as ciliary motion and coughing, airway mucus expels foreign pathogens and toxic nano- and microparticles while selectively allowing the passage of specific nutrients and proteins. These protective and clearance functions depend on the proper rheological properties of mucus under normal physiological conditions. However, in respiratory disease such as CF, COPD, asthma, and COVID-19, excessive mucus secretion is often accompanied by abnormal rheological behaviors. This leads to impaired mucus flow, airway obstruction, and potentially life-threatening conditions. Therefore, this review examines the rheological behaviors of airway mucus in relation to health and disease, focusing on both macrorheology and microrheology. The review highlights those changes in the chemical composition and microstructure of airway mucus, especially under pathological conditions, that can significantly alter its rheological behavior. Rheological parameters can also serve as biological indicators to study the role of mucus in clearance functions and aid in developing pulmonary drug delivery systems. By integrating findings from both macro- and microrheological studies, this review aims to enhance our understanding of the complex behavior of airway mucus, supporting better diagnosis, treatment, and management of chronic respiratory diseases.

physics.med-ph↗

arXiv2025

Multi-Frequency Resonating Based Magnetic Induction Underground Emergency Communications with Diverse Mediums

Magnetic induction (MI) communication is an effective underground emergency communication technique after disasters such as landslides, mine collapses, and earthquakes, due to its advantages in mediums such as soil, concrete, and metals. However, the propagation mediums in practical MI based underground emergency communications are usually diverse and composed randomly due to the impact of disasters, which poses a challenge for MI communication in practical applications. In this paper, we formulate a statistical fading channel model, which reflects the random composition of diverse mediums and is shown to follow a lognormal distribution. To mitigate the impact of diverse medium fading, Multi-frequency Resonating Compensation (MuReC) based coils are used to achieve multiband transmission. Then, we analyze the performance of MuReC based multi-band MI communication with diverse medium fading and derive the expressions of signal-to-noise ratio (SNR) probability density functions, ergodic capacities, average bit error rates (BERs), and outage probabilities for both multiplexing and diversity cases. Numerical results show that MuReC based multiband transmission schemes can effectively reduce the impact of diverse medium fading and enhance the performance.

eess.SP↗

arXiv2025

Robust Beamforming for Magnetic Induction Based Underground Emergency Communications

Magnetic induction (MI) communication is an effective underground emergency communication technique after disasters such as landslides, mine collapses, and earthquakes, due to its advantages in mediums such as soil, concrete, and metals. Based on channel state information (CSI), magnetic beamforming can significantly improve the performance of MI communication. However, in post-disaster underground communication, channel estimation may suffer from errors due to factors such as complex environmental interferences. Taking channel estimation error into account, we formulate a beamforming optimization problem for multi-user MI underground emergency communications, which aims to minimize the power consumption under the constraints of sum rate and signal to interference plus noise ratio (SINR) of each user. Based on the worst-case optimization criterion and the S-procedure, the non-convex optimization problem is transformed into convex and solved. Numerical results show that the proposed robust beamforming scheme can effectively enhance communication reliability and effective throughput in the presence of channel estimation errors.

eess.SP↗

arXiv2025

MiMo: Unlocking the Reasoning Potential of Language Model -- From Pretraining to Posttraining

We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data preprocessing pipeline and employ a three-stage data mixing strategy to strengthen the base model's reasoning potential. MiMo-7B-Base is pre-trained on 25 trillion tokens, with additional Multi-Token Prediction objective for enhanced performance and accelerated inference speed. During post-training, we curate a dataset of 130K verifiable mathematics and programming problems for reinforcement learning, integrating a test-difficulty-driven code-reward scheme to alleviate sparse-reward issues and employing strategic data resampling to stabilize training. Extensive evaluations show that MiMo-7B-Base possesses exceptional reasoning potential, outperforming even much larger 32B models. The final RL-tuned model, MiMo-7B-RL, achieves superior performance on mathematics, code and general reasoning tasks, surpassing the performance of OpenAI o1-mini. The model checkpoints are available at https://github.com/xiaomimimo/MiMo.

cs.CL↗

arXiv2025

MiMo-VL Technical Report

We open-source MiMo-VL-7B-SFT and MiMo-VL-7B-RL, two powerful vision-language models delivering state-of-the-art performance in both general visual understanding and multimodal reasoning. MiMo-VL-7B-RL outperforms Qwen2.5-VL-7B on 35 out of 40 evaluated tasks, and scores 59.4 on OlympiadBench, surpassing models with up to 78B parameters. For GUI grounding applications, it sets a new standard with 56.1 on OSWorld-G, even outperforming specialized models such as UI-TARS. Our training combines four-stage pre-training (2.4 trillion tokens) with Mixed On-policy Reinforcement Learning (MORL) integrating diverse reward signals. We identify the importance of incorporating high-quality reasoning data with long Chain-of-Thought into pre-training stages, and the benefits of mixed RL despite challenges in simultaneous multi-domain optimization. We also contribute a comprehensive evaluation suite covering 50+ tasks to promote reproducibility and advance the field. The model checkpoints and full evaluation suite are available at https://github.com/XiaomiMiMo/MiMo-VL.

cs.CL↗