arXiv subjects

Wang Xu

Publications and source records attributed to Wang Xu.

At least 19 recordsLinked to original sources

arXiv2026

EnvCraft: Synthesizing Executable Environments in Agentic RL for Claw-like Agent

The paradigm of LLMs has rapidly shifted from passive language interfaces to autonomous Claw-like agents that execute long-horizon tasks across stateful workspaces. While Agentic Reinforcement Learning (Agentic RL) provides a promising path to optimize these agents, its scaling is heavily bottlenecked by the severe scarcity of interactive training environments. Existing synthetic environments are strictly limited to tool-calling endpoints, rendering them insufficient for accommodating the end-to-end real-world demands of claw-like agents. To bridge this gap, we introduce EnvCraft, an automated framework for synthesizing executable environments and scalable training data. Specifically, EnvCraft employs an environment synthesis engine to build sandbox-isolated workspaces, alongside a topology-aware data generation engine to produce coherent task trajectories. Overall, we synthesize 139 interactive environments comprising approximately 20K complex tasks for Agentic RL training. Experiments on Qwen3/3.5 models (8B-32B) show that our method yields gains of up to +11.9% on Claw-style benchmarks and +8.0% on general tool-use benchmarks, with concurrent reductions in inference token cost. The results confirm that synthesized executable environments provide robust and generalizable learning signals for training.

cs.AI↗

arXiv2026

SimpleMemVLA: A Simple but Effective Native-Video Memory for Vision-Language-Action Models

Long-horizon manipulation is partially observable: the information needed to choose the next action may appear only in observations from minutes earlier. Existing memory mechanisms: retrieval banks, learned compressors, recurrent states must decide what to keep from the past before knowing what a future decision will require. This was motivated by the assumption that minute-scale history is too large to process directly, which modern VLM backbones no longer make true. In this work, we introduce SimpleMemVLA, a VLA without a dedicated memory module. It keeps the sampled history intact and passes it to the backbone in the timestamped video format the backbone was pretrained to process; the hidden states of a generated sub-task then form the only channel from history to a standard flow-matching action head. Since consecutive decisions share most of their history, prefilling the shared prefix during action execution keeps latency close to a single-frame VLA. SimpleMemVLA sets a new state of the art on four memory benchmarks without cost on general-purpose control. Holding the backbone and training setup fixed, it outperforms retrieval, compression and recurrent-state mechanisms by a wide margin, and causal interventions confirm that the policy genuinely reads its history. Code available at https://github.com/wadeKeith/SimpleMemVLA

cs.CV↗

arXiv2026

DynFly: Dynamic-Aware Continuous Trajectory Generation for UAV Vision-Language Navigation in Urban Environments

Recent advances in multimodal large models have significantly improved UAV vision-language navigation (UAV-VLN) by enhancing high-level perception and reasoning. However, existing methods mainly focus on predicting discrete actions, local targets, or sparse waypoints, while the continuous transition from navigation intent to executable UAV motion remains weakly modeled. This motion-interface gap limits the continuity, stability, and executability of generated UAV trajectories. To address this gap, we propose DynFly, a dynamic-aware continuous trajectory generation framework that bridges high-level navigation reasoning and executable UAV motion. DynFly bridges high-level navigation intent and continuous UAV motion through a lightweight trajectory generation layer. Specifically, it represents expert trajectories in B-spline control-point space and employs a Spline-DiT generator to learn conditional trajectory generation via flow matching. Furthermore, we introduce UAV-oriented dynamic-aware supervision over position, finite-difference velocity, finite-difference acceleration, heading consistency, and local target alignment, enabling the generated trajectories to better satisfy UAV motion characteristics. And our trajectory generation framework can also be integrated with an existing UAV-VLN framework while preserving its original visual-language reasoning pipeline. Extensive experiments on the OpenUAV UAV-VLN benchmark show that DynFly improves both navigation performance and trajectory quality. On the Test Unseen Full split, DynFly improves the strongest baseline by 4.69 NDTW, 2.40 SDTW, 2.14 SR points and 4.87 OSR points, while reducing NE by 4.51 m.

cs.RO↗

arXiv2026

World2VLM: Distilling World Model Imagination into VLMs for Dynamic Spatial Reasoning

Vision-language models (VLMs) have shown strong performance on static visual understanding, yet they still struggle with dynamic spatial reasoning that requires imagining how scenes evolve under egocentric motion. Recent efforts address this limitation either by scaling spatial supervision with synthetic data or by coupling VLMs with world models at inference time. However, the former often lacks explicit modeling of motion-conditioned state transitions, while the latter incurs substantial computational overhead. In this work, we propose World2VLM, a training framework that distills spatial imagination from a generative world model into a vision-language model. Given an initial observation and a parameterized camera trajectory, we use a view-consistent world model to synthesize geometrically aligned future views and derive structured supervision for both forward (action-to-outcome) and inverse (outcome-to-action) spatial reasoning. We post-train the VLM with a two-stage recipe on a compact dataset generated by this pipeline and evaluate it on multiple spatial reasoning benchmarks. World2VLM delivers consistent improvements over the base model across diverse benchmarks, including SAT-Real, SAT-Synthesized, VSI-Bench, and MindCube. It also outperforms the test-time world-model-coupled methods while eliminating the need for expensive inference-time generation. Our results suggest that world models can serve not only as inference-time tools, but also as effective training-time teachers, enabling VLMs to internalize spatial imagination in a scalable and efficient manner.

cs.CV↗

arXiv2026

SpatialFly: Implicit 3D Prior-Guided Visual Reparameterization for Continuous UAV Vision-and-Language Navigation

UAVs play an important role in applications such as autonomous exploration, disaster response, and infrastructure inspection. However, UAV VLN in complex 3D environments remains challenging. A key difficulty is the structural representation mismatch between 2D visual perception and the 3D trajectory decision space, which limits spatial reasoning. To this end, we propose SpatialFly, a geometry-guided spatial representation framework for UAV VLN. Operating on RGB observations without explicit 3D reconstruction, SpatialFly introduces a geometry-guided 2D adaptive representation mechanism. Specifically, the geometric prior injection module injects global structural cues into 2D semantic tokens to provide scene-level geometric guidance. The geometry-aware reparameterization module then uses geometry-conditioned cross-modal attention and gated residual fusion to adaptively reparameterize the visual tokens. Experimental results show that SpatialFly consistently outperforms state-of-the-art UAV VLN baselines across both seen and unseen environments, reducing NE by 4.03m and improving SR by 1.27% over the strongest baseline on the unseen Full split. Additional trajectory-level analysis shows that SpatialFly produces trajectories with better path alignment and smoother, more stable motion.

cs.CV↗

arXiv2026

Cheers: Decoupling Patch Details from Semantic Representations Enables Unified Multimodal Comprehension and Generation

A recent cutting-edge topic in multimodal modeling is to unify visual comprehension and generation within a single model. However, the two tasks demand mismatched decoding regimes and visual representations, making it non-trivial to jointly optimize within a shared feature space. In this work, we present Cheers, a unified multimodal model that decouples patch-level details from semantic representations, thereby stabilizing semantics for multimodal understanding and improving fidelity for image generation via gated detail residuals. Cheers includes three key components: (i) a unified vision tokenizer that encodes and compresses image latent states into semantic tokens for efficient LLM conditioning, (ii) an LLM-based Transformer that unifies autoregressive decoding for text generation and diffusion decoding for image generation, and (iii) a cascaded flow matching head that decodes visual semantics first and then injects semantically gated detail residuals from the vision tokenizer to refine high-frequency content. Experiments on popular benchmarks demonstrate that Cheers matches or surpasses advanced UMMs in both visual understanding and generation. Cheers also achieves 4x token compression, enabling more efficient high-resolution image encoding and generation. Notably, Cheers outperforms the Tar-1.5B on the popular benchmarks GenEval and MMBench, while requiring only 20% of the training cost, indicating effective and efficient (i.e., 4x token compression) unified multimodal modeling. We will release all code and data for future research.

cs.CV↗

arXiv2026

A converse of Berndtsson's theorem on the positivity of direct images

Berndtsson's famous theorem asserts that, for a compact K\"ahler fibration $p:X\to Y$, the direct image bundle $p_*(K_{X/Y}\otimes L)$ of a semi-positive Hermitian holomorphic line bundle $L\to X$ is Nakano semi-positive. As a continuation of our previous work, we prove a converse of Berndtsson's theorem in the case of a projective fibration: if $p_*(K_{X/Y}\otimes L\otimes E)$ is Griffiths semi-positive for every semi-positive Hermitian holomorphic line bundle $E\to X$, then the curvature of $L$ must be semi-positive.

math.CV↗

arXiv2025

MM-UAVBench: How Well Do Multimodal Large Language Models See, Think, and Plan in Low-Altitude UAV Scenarios?

While Multimodal Large Language Models (MLLMs) have exhibited remarkable general intelligence across diverse domains, their potential in low-altitude applications dominated by Unmanned Aerial Vehicles (UAVs) remains largely underexplored. Existing MLLM benchmarks rarely cover the unique challenges of low-altitude scenarios, while UAV-related evaluations mainly focus on specific tasks such as localization or navigation, without a unified evaluation of MLLMs'general intelligence. To bridge this gap, we present MM-UAVBench, a comprehensive benchmark that systematically evaluates MLLMs across three core capability dimensions-perception, cognition, and planning-in low-altitude UAV scenarios. MM-UAVBench comprises 19 sub-tasks with over 5.7K manually annotated questions, all derived from real-world UAV data collected from public datasets. Extensive experiments on 16 open-source and proprietary MLLMs reveal that current models struggle to adapt to the complex visual and cognitive demands of low-altitude scenarios. Our analyses further uncover critical bottlenecks such as spatial bias and multi-view understanding that hinder the effective deployment of MLLMs in UAV scenarios. We hope MM-UAVBench will foster future research on robust and reliable MLLMs for real-world UAV intelligence.

cs.CV↗

arXiv2025

Video2Layout: Recall and Reconstruct Metric-Grounded Cognitive Map for Spatial Reasoning

Spatial intelligence is a critical frontier for Multimodal Large Language Models (MLLMs), empowering them to comprehend the physical world. Drawing inspiration from human perception mechanisms, prior studies attempt to construct a spatial understanding via grid-based cognitive maps. However, current grid-based map methods rely on discretized representations, which limit the model's ability in fine-grained spatial reasoning. To overcome this limitation, we propose Video2Layout, a framework for reconstructing metric-grounded spatial layouts from video. The framework uses continuous object boundary coordinates to enable quantitative spatial computation, which effectively reduces ambiguity in natural language descriptions of spatial relationships. Specifically, our method comprises two stages. First, in supervised fine-tuning stage, we construct a high-quality dataset from the AI2THOR simulator, which enables the model to learn the mapping from visual inputs to precise boundary coordinates. Subsequently, a reinforcement fine-tuning stage enhances the model's real-world generalization capabilities. Based on the above framework, we investigate factors that affect cognitive map accuracy and quantify its relationship with task performance. Evaluated on mainstream spatial reasoning benchmarks, our model, V2LO-7B, achieves an average improvement of 3.24\% over the model trained on grid maps, validating the superiority of our method.

cs.CV↗

arXiv2025

Electrochemical properties of solid oxide fuel cells under the coupling effect of airflow pattern and airflow velocity

Under the dual background of deep adjustment of global energy pattern and severe challenges of environmental problems, solid oxide fuel cell (SOFC) has become the focus of research on efficient and clean energy conversion technology due to its many excellent characteristics. The electrochemical performance of SOFC is affected by various factors such as gas flow pattern (co-flow, counter-flow, cross-flow), flow rate (cathode and anode channel gases), and operating voltage. Accurately analysing the variation of electrochemical indexes with each factor is the basis for proposing the design scheme of high efficiency reaction of the cell. Therefore, a three-dimensional multi-field coupling model of SOFC is established in this study, and the model parameters and boundary conditions covering electrochemistry, gas flow, substance diffusion, etc. are set to study the influence of the coupling between factors on the electrochemical performance of the cell. These results show that with the decrease of operating voltage, the electrochemical reaction rate of the cell increases significantly, the gas mole fraction gradient increases, and the inhomogeneity of the electrolyte current density distribution is enhanced. Under low-voltage operating conditions, the cross-flow flow pattern shows better electrochemical performance advantages, and its power density profile takes the lead in different current density intervals. With the increase of the flow rate of the flow channel gas, the output power density curve of the cell shows an overall upward trend, and then the driving effect of the flow rate increase on the power density increase is gradually weakened due to the saturated cathodic reaction. This study reveals the influence of the coupling of flow pattern, flow rate and voltage on the electrochemical performance of SOFC, and provides guidance for the commercial application of SOFC.

physics.app-ph↗

arXiv2025

DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models

Does Chain-of-Thought (CoT) reasoning genuinely improve Vision Language Action (VLA) models, or does it merely add overhead? Existing CoT-VLA systems report limited and inconsistent gains, yet no prior work has rigorously diagnosed when and why CoT helps robots act. Through systematic experiments, we identify two necessary conditions that must be jointly satisfied for CoT to be effective in VLA: (1) Decoding Alignment: CoT and actions must be generated with modality-appropriate mechanisms; forcing both through a single autoregressive decoder is not merely suboptimal but actively harmful, degrading performance by 4.2 percentage points; (2) Causal Alignment: CoT must be causally linked to task success via outcome-based optimization; without it, supervised CoT is indistinguishable from no reasoning at all under action-execution-sensitive dynamics shift, exhibiting a 32.0 pp performance drop nearly identical to the 31.6 pp drop of a reasoning-free baseline. Guided by these findings, we build DeepThinkVLA: a hybrid-attention decoder satisfies Condition 1 by pairing causal attention for language with bidirectional attention for parallel action decoding, while a two-stage SFT-then-RL pipeline satisfies Condition 2 by aligning the full reasoning: action chain with sparse task-success rewards. DeepThinkVLA achieves 97.0\% success on LIBERO, 79.0\% robustness on LIBERO-Plus (vs. 61.6\% for $\pi_0$-FAST), and 59.3\% success on RoboTwin 2.0, exceeding the strongest baseline by 21.7 points. Furthermore, real-robot experiments provide preliminary evidence for the physical applicability of our CoT data construction and hybrid architecture. Our codes are available at https://github.com/OpenBMB/DeepThinkVLA.

cs.LG↗

arXiv2025

Why Do MLLMs Struggle with Spatial Understanding? A Systematic Analysis from Data to Architecture

Spatial understanding is essential for Multimodal Large Language Models (MLLMs) to support perception, reasoning, and planning in embodied environments. Despite recent progress, existing studies reveal that MLLMs still struggle with spatial understanding. However, existing research lacks a comprehensive and systematic evaluation of these limitations, often restricted to isolated scenarios, such as single-view or video. In this work, we present a systematic analysis of spatial understanding from both data and architectural perspectives across three representative scenarios: single-view, multi-view, and video. We propose a benchmark named MulSeT (Multi-view Spatial Understanding Tasks), and design a series of experiments to analyze the spatial reasoning capabilities of MLLMs. From the data perspective, the performance of spatial understanding converges quickly as the training data increases, and the upper bound is relatively low, especially for tasks that require spatial imagination. This indicates that merely expanding training data is insufficient to achieve satisfactory performance. From the architectural perspective, we find that spatial understanding relies more heavily on the positional encoding within the visual encoder than within the language model, in both cascaded and native MLLMs. Moreover, we explore reasoning injection and envision future improvements through architectural design to optimize spatial understanding. These insights shed light on the limitations of current MLLMs and suggest new directions for improving spatial reasoning capabilities through data scaling and architectural tuning.

cs.CV↗

arXiv2025

CookBench: A Long-Horizon Embodied Planning Benchmark for Complex Cooking Scenarios

Embodied Planning is dedicated to the goal of creating agents capable of executing long-horizon tasks in complex physical worlds. However, existing embodied planning benchmarks frequently feature short-horizon tasks and coarse-grained action primitives. To address this challenge, we introduce CookBench, a benchmark for long-horizon planning in complex cooking scenarios. By leveraging a high-fidelity simulation environment built upon the powerful Unity game engine, we define frontier AI challenges in a complex, realistic environment. The core task in CookBench is designed as a two-stage process. First, in Intention Recognition, an agent needs to accurately parse a user's complex intent. Second, in Embodied Interaction, the agent should execute the identified cooking goal through a long-horizon, fine-grained sequence of physical actions. Unlike existing embodied planning benchmarks, we refine the action granularity to a spatial level that considers crucial operational information while abstracting away low-level robotic control. Besides, We provide a comprehensive toolset that encapsulates the simulator. Its unified API supports both macro-level operations, such as placing orders and purchasing ingredients, and a rich set of fine-grained embodied actions for physical interaction, enabling researchers to focus on high-level planning and decision-making. Furthermore, we present an in-depth analysis of state-of-the-art, closed-source Large Language Model and Vision-Language Model, revealing their major shortcomings and challenges posed by complex, long-horizon tasks. The full benchmark will be open-sourced to facilitate future research.

cs.RO↗

arXiv2025

EfficientEdit: Accelerating Code Editing via Edit-Oriented Speculative Decoding

Large Language Models (LLMs) have demonstrated remarkable capabilities in code editing, substantially enhancing software development productivity. However, the inherent complexity of code editing tasks forces existing approaches to rely on LLMs' autoregressive end-to-end generation, where decoding speed plays a critical role in efficiency. While inference acceleration techniques like speculative decoding are applied to improve the decoding efficiency, these methods fail to account for the unique characteristics of code editing tasks where changes are typically localized and existing code segments are reused. To address this limitation, we propose EfficientEdit, a novel method that improves LLM-based code editing efficiency through two key mechanisms based on speculative decoding: (1) effective reuse of original code segments while identifying potential edit locations, and (2) efficient generate edit content via high-quality drafts from edit-oriented draft models and a dynamic verification mechanism that balances quality and acceleration. Experimental results show that EfficientEdit can achieve up to 10.38$\times$ and 13.09$\times$ speedup compared to standard autoregressive decoding in CanItEdit and CodeIF-Bench, respectively, outperforming state-of-the-art inference acceleration approaches by up to 90.6%. The code and data are available at https://github.com/zhu-zhu-ding/EfficientEdit.

cs.SE↗

arXiv2025

AgentCPM-GUI: Building Mobile-Use Agents with Reinforcement Fine-Tuning

The recent progress of large language model agents has opened new possibilities for automating tasks through graphical user interfaces (GUIs), especially in mobile environments where intelligent interaction can greatly enhance usability. However, practical deployment of such agents remains constrained by several key challenges. Existing training data is often noisy and lack semantic diversity, which hinders the learning of precise grounding and planning. Models trained purely by imitation tend to overfit to seen interface patterns and fail to generalize in unfamiliar scenarios. Moreover, most prior work focuses on English interfaces while overlooks the growing diversity of non-English applications such as those in the Chinese mobile ecosystem. In this work, we present AgentCPM-GUI, an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction. Our training pipeline includes grounding-aware pre-training to enhance perception, supervised fine-tuning on high-quality Chinese and English trajectories to imitate human-like actions, and reinforcement fine-tuning with GRPO to improve reasoning capability. We also introduce a compact action space that reduces output length and supports low-latency execution on mobile devices. AgentCPM-GUI achieves state-of-the-art performance on five public benchmarks and a new Chinese GUI benchmark called CAGUI, reaching $96.9\%$ Type-Match and $91.3\%$ Exact-Match. To facilitate reproducibility and further research, we publicly release all code, model checkpoint, and evaluation data.

cs.AI↗

arXiv2025

Speculative Decoding Meets Quantization: Compatibility Evaluation and Hierarchical Framework Design

Speculative decoding and quantization effectively accelerate memory-bound inference of large language models. Speculative decoding mitigates the memory bandwidth bottleneck by verifying multiple tokens within a single forward pass, which increases computational effort. Quantization achieves this optimization by compressing weights and activations into lower bit-widths and also reduces computations via low-bit matrix multiplications. To further leverage their strengths, we investigate the integration of these two techniques. Surprisingly, experiments applying the advanced speculative decoding method EAGLE-2 to various quantized models reveal that the memory benefits from 4-bit weight quantization are diminished by the computational load from speculative decoding. Specifically, verifying a tree-style draft incurs significantly more time overhead than a single-token forward pass on 4-bit weight quantized models. This finding led to our new speculative decoding design: a hierarchical framework that employs a small model as an intermediate stage to turn tree-style drafts into sequence drafts, leveraging the memory access benefits of the target quantized model. Experimental results show that our hierarchical approach achieves a 2.78$\times$ speedup across various tasks for the 4-bit weight Llama-3-70B model on an A100 GPU, outperforming EAGLE-2 by 1.31$\times$. Code available at https://github.com/AI9Stars/SpecMQuant.

cs.CL↗

arXiv2025

Learning-based 3D Reconstruction in Autonomous Driving: A Comprehensive Survey

Learning-based 3D reconstruction has emerged as a transformative technique in autonomous driving, enabling precise modeling of environments through advanced neural representations. It has inspired pioneering solutions for vital tasks in autonomous driving, such as dense mapping and closed-loop simulation, as well as comprehensive scene feature for driving scene understanding and reasoning. Given the rapid growth in related research, this survey provides a comprehensive review of both technical evolutions and practical applications in autonomous driving. We begin with an introduction to the preliminaries of learning-based 3D reconstruction to provide a solid technical background foundation, then progress to a rigorous, multi-dimensional examination of cutting-edge methodologies, systematically organized according to the distinctive technical requirements and fundamental challenges of autonomous driving. Through analyzing and summarizing development trends and cutting-edge research, we identify existing technical challenges, along with insufficient disclosure of on-board validation and safety verification details in the current literature, and ultimately suggest potential directions to guide future studies.

cs.CV↗

arXiv2025

DuplexMamba: Enhancing Real-time Speech Conversations with Duplex and Streaming Capabilities

Real-time speech conversation is essential for natural and efficient human-machine interactions, requiring duplex and streaming capabilities. Traditional Transformer-based conversational chatbots operate in a turn-based manner and exhibit quadratic computational complexity that grows as the input size increases. In this paper, we propose DuplexMamba, a Mamba-based end-to-end multimodal duplex model for speech-to-text conversation. DuplexMamba enables simultaneous input processing and output generation, dynamically adjusting to support real-time streaming. Specifically, we develop a Mamba-based speech encoder and adapt it with a Mamba-based language model. Furthermore, we introduce a novel duplex decoding strategy that enables DuplexMamba to process input and generate output simultaneously. Experimental results demonstrate that DuplexMamba successfully implements duplex and streaming capabilities while achieving performance comparable to several recently developed Transformer-based models in automatic speech recognition (ASR) tasks and voice assistant benchmark evaluations. Our code and model are released.

cs.CL↗