arXiv subjects

Yuyang Chen

Publications and source records attributed to Yuyang Chen.

At least 19 recordsLinked to original sources

arXiv2026

VLASS Discovery of a Luminous Galactic Radio Transient Evolving on Decade Timescales

We present a multiwavelength analysis of the radio transient VT J1906+0849, discovered as a 70 mJy source in Epoch 1 of the Very Large Array Sky Survey (VLASS), 21 yr after an NRAO VLA Sky Survey (NVSS) non-detection. Radio observations reveal the source was first detected in 2005, peaking at $\gtrsim200$ mJy in 2014, then declining until a late 2025 rebrightening. The transient sits at a Galactic latitude of $\approx0.74^\circ$ and the properties of its optical-infrared counterpart support a Galactic origin. At $d\gtrsim15$ kpc, the extreme radio luminosity is likely powered by sustained accretion onto a compact object. However, a Swift-XRT non-detection shows it is X-ray faint relative to the Galactic X-ray binary population, and the radio emission is distinct from X-ray binaries in its temporal and spectral behavior. Broadband radio spectra suggest synchrotron self-absorption, but size constraints from equipartition and very long baseline interferometry show little to no expansion in the radio-emitting region over 5+ yr, despite significant spectral evolution. Near-infrared spectroscopy reveals a single broad emission line with a stable centroid but variable width and luminosity. We attribute this feature to blueshifted Br$\gamma$ tracing a persistent asymmetric $\approx2000$ km s$^{-1}$ outflow. These properties are unlike any previously identified Galactic radio source. One possible interpretation is that VT J1906+0849 is a young analog of the microquasar SS 433, with a dense disk wind confining a continuously powered synchrotron outflow. This jet-wind interaction explains the compact, slowly expanding radio source and may contribute to the absence of bright X-ray emission.

astro-ph.HE↗

arXiv2026

DiTango: Cost-Effective Parallel Diffusion Generation with Selective Attention State Reuse

Recent advances in AI-generated content have driven widespread adoption of Diffusion Transformers (DiTs) for high-resolution, long-duration content generation. While parallelization techniques accelerate diffusion inference, they face significant scalability challenges due to excessive communication overhead in multi-node environments. We observe that sequence partitions in Context Parallelism (CP) exhibit distinct heterogeneity: spatially proximate partitions contribute more significantly to attention computation results. By mapping this heterogeneous pattern to hierarchical communication topology, we can access high-contribution partitions with reduced communication cost. This insight motivates our novel selective attention state mechanism that strategically balances partial attention computation and historical result reuse across denoising steps. We present DiTango, an efficient parallel framework for DiT generation. DiTango features an anchor-guided state selection planner that optimizes computation-reuse decisions for each partition, complemented by a runtime that orchestrates efficient state-centric operations. This design achieves superior system efficiency while preserving generation quality. Experimental evaluation on popular diffusion models demonstrates that DiTango achieves up to 1.9x end-to-end and 3.2x attention speedup with near-linear scaling in multi-node settings, while maintaining generation quality comparable to state-of-the-art approaches.

cs.CV↗

arXiv2026

WebRetriever: A Large-Scale Comprehensive Benchmark for Efficient Web Agent Evaluation

As web agents increasingly demonstrate capabilities in automated task execution, the development of robust evaluation frameworks for assessing their navigation and task completion performance has emerged as a critical research priority. However, existing benchmarks exhibit fundamental limitations. First, they suffer from insufficient scale and limited domain diversity, constraining comprehensive evaluation of cross-domain generalization. Second, prevailing LLM-as-Judge evaluation methodologies inadequately capture fine-grained interaction semantics, particularly regarding precise query formulation and filtering operations. Third, current benchmarks predominantly emphasize navigation success metrics while neglecting critical requirements for real-world deployment scenarios. To address these limitations, we introduce WebRetriever, a large-scale benchmark encompassing 800 websites and 1,550 tasks across diverse domains, including consumer, professional, and enterprise sectors, with comprehensive coverage of user intent patterns. We propose NavEval (Navigation Evaluation), a novel LLM-as-Judge framework that leverages rich interaction context beyond visual screenshots, achieving state-of-the-art alignment with human judgment across multiple evaluation datasets. Furthermore, we establish three complementary evaluation protocols that collectively provide holistic assessment of web agent capabilities: navigation proficiency, knowledge-assisted interaction, and end-to-end task completion with information extraction. Extensive experimental analysis reveals substantial performance disparities across evaluation protocols, demonstrating that navigation success alone is an insufficient predictor of real-world application effectiveness. WebRetriever delivers fine-grained diagnostic insights into agent capabilities and establishes a rigorous foundation for advancing web agent research and development.

cs.CV↗

arXiv2026

E$^2$DT: Efficient and Effective Decision Transformer with Experience-Aware Sampling for Robotic Manipulation

In reinforcement learning (RL) for robotic manipulation, the Decision Transformer (DT) has emerged as an effective framework for addressing long-horizon tasks. However, DT's performance depends heavily on the coverage of collected experiences. Without an active exploration mechanism, standard DT relies on uniform replay, which leads to poor sample efficiency, limited exploration, and reduced overall effectiveness. At the same time, while excessive exploration can help avoid local optima, it often delays policy convergence and leads to degraded efficiency. To address these limitations, we propose E$^2$DT, a DT-guided k-Determinantal Point Process sampling framework that enables the model to actively shape its own experience selection. Our framework is experience-aware, allowing E$^2$DT to be both efficient, by prioritizing sampling quality, such as high-return, high-uncertainty, and underrepresented trajectories, and effective, by ensuring diversity across trajectory windows to preserve policy optimality. Specifically, DT's internal latent embeddings measure diversity across trajectory windows, while quality is quantified through a composite metric that integrates return-to-go (RTG) quantiles, predictive uncertainty, and stage coverage based on inverse frequency. These two dimensions are integrated into a novel quality-diversity joint kernel that prioritizes the most informative experiences, thereby enabling learning that is both efficient and effective. We evaluate E$^2$DT on challenging robotic manipulation benchmarks in both simulation and real-robot settings. Results show that it consistently outperforms prior methods. These findings demonstrate that coupling policy learning with experience-aware sampling provides a principled path toward robust long-horizon robotic learning.

cs.RO↗

arXiv2026

C$^2$T: Captioning-Structure and LLM-Aligned Common-Sense Reward Learning for Traffic--Vehicle Coordination

State-of-the-art (SOTA) urban traffic control increasingly employs Multi-Agent Reinforcement Learning (MARL) to coordinate Traffic Light Controllers (TLCs) and Connected Autonomous Vehicles (CAVs). However, the performance of these systems is fundamentally capped by their hand-crafted, myopic rewards (e.g., intersection pressure), which fail to capture high-level, human-centric goals like safety, flow stability, and comfort. To overcome this limitation, we introduce C2T, a novel framework that learns a common-sense coordination model from traffic-vehicle dynamics. C2T distills "common-sense" knowledge from a Large Language Model (LLM) into a learned intrinsic reward function. This new reward is then used to guide the coordination policy of a cooperative multi-intersection TLC MARL system on CityFlow-based multi-intersection benchmarks. Our framework significantly outperforms strong MARL baselines in traffic efficiency, safety, and an energy-related proxy. We further highlight C2T's flexibility in principle, allowing distinct "efficiency-focused" versus "safety-focused" policies by modifying the LLM prompt.

cs.MA↗

arXiv2026

Diffusion-based Probabilistic Air Quality Forecasting with Mechanistic Insight

Current operational air quality forecasts are computationally expensive, sensitive to errors in physics and emissions, and often neglect weather-related uncertainty. To address these limitations, we present AirFusion, a hybrid, diffusion-based framework that synergistically integrates knowledge from chemical transport models with real-world observational constraints to enable accurate and efficient probabilistic regional air quality prediction. We apply AirFusion to generate operational 6-day, 30-member ensemble forecasts of surface ozone across China, initialized with observations and driven by ensemble weather forecasts. AirFusion outperforms existing operational benchmarks, achieving substantially lower forecast errors against surface measurements, while also providing ensemble-based diagnostics that explicitly quantify the impacts of weather uncertainty on air quality predictability. Moreover, AirFusion can rapidly adapt to evolving emissions through fine-tuning with only one month of recent observations. These attributes establish AirFusion as a powerful and extensible framework for next-generation probabilistic air quality forecasting, with clear potential for application to other pollutants and regions.

physics.ao-ph↗

arXiv2026

Jano: Adaptive Diffusion Generation with Early-stage Convergence Awareness

Diffusion models have achieved remarkable success in generative AI, yet their computational efficiency remains a significant challenge, particularly for Diffusion Transformers (DiTs) requiring intensive full-attention computation. While existing acceleration approaches focus on content-agnostic uniform optimization strategies, we observe that different regions in generated content exhibit heterogeneous convergence patterns during the denoising process. We present Jano, a training-free framework that leverages this insight for efficient region-aware generation. Jano introduces an early-stage complexity recognition algorithm that accurately identifies regional convergence requirements within initial denoising steps, coupled with an adaptive token scheduling runtime that optimizes computational resource allocation. Through comprehensive evaluation on state-of-the-art models, Jano achieves substantial acceleration (average 2.0 times speedup, up to 2.4 times) while preserving generation quality. Our work challenges conventional uniform processing assumptions and provides a practical solution for accelerating large-scale content generation. The source code of our implementation is available at https://github.com/chen-yy20/Jano.

cs.CV↗

arXiv2026

DM0: An Embodied-Native Vision-Language-Action Model towards Physical AI

Moving beyond the traditional paradigm of adapting internet-pretrained models to physical tasks, we present DM0, an Embodied-Native Vision-Language-Action (VLA) framework designed for Physical AI. Unlike approaches that treat physical grounding as a fine-tuning afterthought, DM0 unifies embodied manipulation and navigation by learning from heterogeneous data sources from the onset. Our methodology follows a comprehensive three-stage pipeline: Pretraining, Mid-Training, and Post-Training. First, we conduct large-scale unified pretraining on the Vision-Language Model (VLM) using diverse corpora--seamlessly integrating web text, autonomous driving scenarios, and embodied interaction logs-to jointly acquire semantic knowledge and physical priors. Subsequently, we build a flow-matching action expert atop the VLM. To reconcile high-level reasoning with low-level control, DM0 employs a hybrid training strategy: for embodied data, gradients from the action expert are not backpropagated to the VLM to preserve generalized representations, while the VLM remains trainable on non-embodied data. Furthermore, we introduce an Embodied Spatial Scaffolding strategy to construct spatial Chain-of-Thought (CoT) reasoning, effectively constraining the action solution space. Experiments on the RoboChallenge benchmark demonstrate that DM0 achieves state-of-the-art performance in both Specialist and Generalist settings on Table30.

cs.RO↗

arXiv2026

STEP3-VL-10B Technical Report

We present STEP3-VL-10B, a lightweight open-source foundation model designed to redefine the trade-off between compact efficiency and frontier-level multimodal intelligence. STEP3-VL-10B is realized through two strategic shifts: first, a unified, fully unfrozen pre-training strategy on 1.2T multimodal tokens that integrates a language-aligned Perception Encoder with a Qwen3-8B decoder to establish intrinsic vision-language synergy; and second, a scaled post-training pipeline featuring over 1k iterations of reinforcement learning. Crucially, we implement Parallel Coordinated Reasoning (PaCoRe) to scale test-time compute, allocating resources to scalable perceptual reasoning that explores and synthesizes diverse visual hypotheses. Consequently, despite its compact 10B footprint, STEP3-VL-10B rivals or surpasses models 10$\times$-20$\times$ larger (e.g., GLM-4.6V-106B, Qwen3-VL-235B) and top-tier proprietary flagships like Gemini 2.5 Pro and Seed-1.5-VL. Delivering best-in-class performance, it records 92.2% on MMBench and 80.11% on MMMU, while excelling in complex reasoning with 94.43% on AIME2025 and 75.95% on MathVision. We release the full model suite to provide the community with a powerful, efficient, and reproducible baseline.

cs.CV↗

arXiv2025

Mano Technical Report

Graphical user interfaces (GUIs) are the primary medium for human-computer interaction, yet automating GUI interactions remains challenging due to the complexity of visual elements, dynamic environments, and the need for multi-step reasoning. Existing methods based on vision-language models (VLMs) often suffer from limited resolution, domain mismatch, and insufficient sequential decisionmaking capability. To address these issues, we propose Mano, a robust GUI agent built upon a multi-modal foundation model pre-trained on extensive web and computer system data. Our approach integrates a novel simulated environment for high-fidelity data generation, a three-stage training pipeline (supervised fine-tuning, offline reinforcement learning, and online reinforcement learning), and a verification module for error recovery. Mano demonstrates state-of-the-art performance on multiple GUI benchmarks, including Mind2Web and OSWorld, achieving significant improvements in success rate and operational accuracy. Our work provides new insights into the effective integration of reinforcement learning with VLMs for practical GUI agent deployment, highlighting the importance of domain-specific data, iterative training, and holistic reward design.

cs.MM↗

arXiv2025

HARDMath2: A Benchmark for Applied Mathematics Built by Students as Part of a Graduate Class

Large language models (LLMs) have shown remarkable progress in mathematical problem-solving, but evaluation has largely focused on problems that have exact analytical solutions or involve formal proofs, often overlooking approximation-based problems ubiquitous in applied science and engineering. To fill this gap, we build on prior work and present HARDMath2, a dataset of 211 original problems covering the core topics in an introductory graduate applied math class, including boundary-layer analysis, WKB methods, asymptotic solutions of nonlinear partial differential equations, and the asymptotics of oscillatory integrals. This dataset was designed and verified by the students and instructors of a core graduate applied mathematics course at Harvard. We build the dataset through a novel collaborative environment that challenges students to write and refine difficult problems consistent with the class syllabus, peer-validate solutions, test different models, and automatically check LLM-generated solutions against their own answers and numerical ground truths. Evaluation results show that leading frontier models still struggle with many of the problems in the dataset, highlighting a gap in the mathematical reasoning skills of current LLMs. Importantly, students identified strategies to create increasingly difficult problems by interacting with the models and exploiting common failure modes. This back-and-forth with the models not only resulted in a richer and more challenging benchmark but also led to qualitative improvements in the students' understanding of the course material, which is increasingly important as we enter an age where state-of-the-art language models can solve many challenging problems across a wide domain of fields.

cs.LG↗

arXiv2025

Krysalis Hand: A Lightweight, High-Payload, 18-DoF Anthropomorphic End-Effector for Robotic Learning and Dexterous Manipulation

This paper presents the Krysalis Hand, a five-finger robotic end-effector that combines a lightweight design, high payload capacity, and a high number of degrees of freedom (DoF) to enable dexterous manipulation in both industrial and research settings. This design integrates the actuators within the hand while maintaining an anthropomorphic form. Each finger joint features a self-locking mechanism that allows the hand to sustain large external forces without active motor engagement. This approach shifts the payload limitation from the motor strength to the mechanical strength of the hand, allowing the use of smaller, more cost-effective motors. With 18 DoF and weighing only 790 grams, the Krysalis Hand delivers an active squeezing force of 10 N per finger and supports a passive payload capacity exceeding 10 lbs. These characteristics make Krysalis Hand one of the lightest, strongest, and most dexterous robotic end-effectors of its kind. Experimental evaluations validate its ability to perform intricate manipulation tasks and handle heavy payloads, underscoring its potential for industrial applications as well as academic research. All code related to the Krysalis Hand, including control and teleoperation, is available on the project GitHub repository: https://github.com/Soltanilara/Krysalis_Hand

cs.RO↗

arXiv2025

Enhancing LLM Agents for Code Generation with Possibility and Pass-rate Prioritized Experience Replay

Nowadays transformer-based Large Language Models (LLM) for code generation tasks usually apply sampling and filtering pipelines. Due to the sparse reward problem in code generation tasks caused by one-token incorrectness, transformer-based models will sample redundant programs till they find a correct one, leading to low efficiency. To overcome the challenge, we incorporate Experience Replay (ER) in the fine-tuning phase, where codes and programs produced are stored and will be replayed to give the LLM agent a chance to learn from past experiences. Based on the spirit of ER, we introduce a novel approach called BTP pipeline which consists of three phases: beam search sampling, testing phase, and prioritized experience replay phase. The approach makes use of failed programs collected by code models and replays programs with high Possibility and Pass-rate Prioritized value (P2Value) from the replay buffer to improve efficiency. P2Value comprehensively considers the possibility of transformers' output and pass rate and can make use of the redundant resources caused by the problem that most programs collected by LLMs fail to pass any tests. We empirically apply our approach in several LLMs, demonstrating that it enhances their performance in code generation tasks and surpasses existing baselines.

cs.LG↗

arXiv2024

Quadrupole topological behavior of elastic waves in two-dimensional square lattices with nonsymmorphic symmetries

We investigate a novel higher-order topological behavior in elastic lattices characterized by nonsymmorphic symmetries. In the theoretical spring-mass lattice, altering the vertex mass allows for fine-tuning of the topological features within the bandgap. We analyze the quadrupole topological behavior in square lattices with nonsymmorphic symmetries using nested Wannier bands. Beyond second-order topological metamaterials, a single-phase topological configuration promotes energy localization at the corners due to a non-zero relative quadrupole moment. Our findings are validated through experimental observations of higher-order topological corner states, which show excellent agreement with simulated results and theoretical predictions. Additionally, the elastic lattices in the self-similar system exhibit fractal higher-order topological behaviors, revealing numerous topological edge and corner states. The self-similar lattice also demonstrates enhanced energy localization, with the number of topological states showing a linear correlation to the corner dimension. This study provides new insights into elastic higher-order topological insulators and inspires innovative strategies for simulating topological elastic materials.

physics.app-ph↗

arXiv2024

Dynamic Attention and Bi-directional Fusion for Safety Helmet Wearing Detection

Ensuring construction site safety requires accurate and real-time detection of workers' safety helmet use, despite challenges posed by cluttered environments, densely populated work areas, and hard-to-detect small or overlapping objects caused by building obstructions. This paper proposes a novel algorithm for safety helmet wearing detection, incorporating a dynamic attention within the detection head to enhance multi-scale perception. The mechanism combines feature-level attention for scale adaptation, spatial attention for spatial localization, and channel attention for task-specific insights, improving small object detection without additional computational overhead. Furthermore, a two-way fusion strategy enables bidirectional information flow, refining feature fusion through adaptive multi-scale weighting, and enhancing recognition of occluded targets. Experimental results demonstrate a 1.7% improvement in mAP@[.5:.95] compared to the best baseline while reducing GFLOPs by 11.9% on larger sizes. The proposed method surpasses existing models, providing an efficient and practical solution for real-world construction safety monitoring.

cs.CV↗

arXiv2024

Efficient Diversity-based Experience Replay for Deep Reinforcement Learning

Experience replay is widely used to improve learning efficiency in reinforcement learning by leveraging past experiences. However, existing experience replay methods, whether based on uniform or prioritized sampling, often suffer from low efficiency, particularly in real-world scenarios with high-dimensional state spaces. To address this limitation, we propose a novel approach, Efficient Diversity-based Experience Replay (EDER). EDER employs a determinantal point process to model the diversity between samples and prioritizes replay based on the diversity between samples. To further enhance learning efficiency, we incorporate Cholesky decomposition for handling large state spaces in realistic environments. Additionally, rejection sampling is applied to select samples with higher diversity, thereby improving overall learning efficacy. Extensive experiments are conducted on robotic manipulation tasks in MuJoCo, Atari games, and realistic indoor environments in Habitat. The results demonstrate that our approach not only significantly improves learning efficiency but also achieves superior performance in high-dimensional, realistic environments.

cs.LG↗

arXiv2024

Searching for Radio Transients with Inverted Spectra in Epoch 1 of VLASS and VCSS, and Identification of a Sample of Candidate Relativistic Nuclear Transients

For radio transients, an inverted spectrum (defined as $\alpha > 0$ for a power law spectrum $S_{\nu}\propto \nu^{\alpha}$) constrains physical properties, which in principle can be a useful criterion for selecting specific targets of interest in a transient search. To test and develop this concept, we have searched epoch 1 of the Very Large Array Sky Survey (VLASS; 3.0 GHz) and the VLITE Commensal Sky Survey (VCSS; 340 MHz) for radio transients with inverted spectra. We discover a sample of 21 inverted-spectra transient candidates that are not associated with cataloged active galactic nuclei (AGNs). To the best of our knowledge, three of our candidates have recently been reported by others as radio transients, but none have reported transient counterparts at other wavelengths. We find that our candidates evolve slowly over years and show either highly inverted spectra or peaked spectra over $\sim$1--3 GHz. Within our sample, nine candidates are matched to optical centers of galaxies and have estimated radio spectral luminosities of $L_{3.0\mathrm{GHz}}\sim10^{30}-10^{33}\,\mathrm{erg}\,\mathrm{s}^{-1}\,\mathrm{Hz}^{-1}$. Based on the observed properties, we find the most plausible transient classification for our candidates to be relativistic tidal disruption events. However, it is difficult to rule out variable or transient AGNs with highly inverted spectra. Upon examining physical constraints, we confirm that mainly relativistic transients (on-axis or off-axis) with equipartition energy $E_{\mathrm{eq}}\gtrsim10^{49}-10^{53}\,\mathrm{erg}$ are expected from searching VLASS and VCSS based on inverted spectra. The obtainable physical constraints, however, can be weak due to degeneracy introduced by viewing angle.

astro-ph.HE↗

arXiv2024

A robust stochastic control problem with applications to monotone mean-variance problems

This paper studies a robust stochastic control problem with a monotone mean-variance cost functional and random coefficients. The main technique is to find the saddle point through two backward stochastic differential equations (BSDEs) with unbounded coefficients. We further show that the robust stochastic control problem shares the same optimal control and optimal value with the stochastic control problem with a mean-variance cost functional. The results obtained are then applied to monotone mean-variance and mean-variance portfolio selection problems and monotone mean-variance and mean-variance investment-reinsurance problems.

math.OC↗