arXiv subjects

Xinyi Wang

Publications and source records attributed to Xinyi Wang.

At least 19 recordsLinked to original sources

arXiv2026

TensorSSM: Memorizing Spatial Position in Dynamic State Space for Tensor Core Generation

High-dimensional tensors are ubiquitous in real-world applications but lead to massive storage overhead. Functional tensor decomposition addresses this by employing neural networks to map element coordinates to tensor cores, significantly reducing the parameters for data compression. However, existing functional methods typically flatten high-order tensors into 1D sequences. This folding operation causes highly overlapping local coordinate sequences and struggles to capture complex cross-mode spatial dependencies, severely limiting their expressiveness on diverse datasets. To tackle these challenges, we propose TensorSSM, a novel Implicit Neural Representation (INR) for functional decomposition method that memorizes spatial positions within a dynamic State Space Model (SSM). Specifically, TensorSSM first decomposes mode-wise coordinates into sequences spit by K-ary rule, encoding spatial tokens from coarse to fine, which effectively captures spatial variations. Furthermore, we extend the SSM's state space to explicitly map cross-mode spatial dependencies, allowing inter-mode interactions to enhance the updates of each mode. By integrating cross-mode information from previous hidden states into the current state transitions, TensorSSM effectively expands tensor representation along two dimensions: the sequence dimension for coarse-to-fine spatial tokens, and the state dimension for cross-mode memorization. This design effectively integrates spatial information into the state space and the tensor core, substantially advancing the state adaptability and enhancing the expressiveness of the data compression. Extensive experiments demonstrate that TensorSSM achieves significantly high reconstruction accuracy and storage efficiency, exhibiting robust expressiveness across diverse tensor datasets without relying on strict spatial assumptions.

cs.LG↗

arXiv2026

In-Orbit Computing for Remote Sensing: Task-Oriented Processing and Transmission

The growing volume and timeliness requirements of remote sensing data are exposing the limitations of the conventional "transmit-then-process" architecture, particularly under intermittent satellite-ground connectivity, constrained downlink bandwidth, and limited onboard computing resources. This article presents a semantic-aware space-ground collaborative remote sensing (SSGC-RS) framework, in which remote sensing data are processed to different semantic depths according to task requirements and available communication-computing resources. Three operating modes are considered, ranging from fully onboard task inference to semantic representation delivery and space-ground split inference. A land-cover recognition case study illustrates how these modes address different communication and computing bottlenecks. Finally, emerging challenges and future directions for intelligent remote sensing systems are discussed.

eess.SP↗

arXiv2026

Echo: Coupled Action and Task Status Signal Denoising for Vision Language Action Models

Recognizing task completion and failure enables robots to stop upon success and initiate recovery or replanning after failure. Existing methods augment Vision Language Action (VLA) policies with task status signal prediction to support these execution decisions. However, learning and integrating these signals can require fine-grained supervision, external monitoring models, or complex architectural designs. Moreover, the potential of task status signals to directly guide action generation remains underexplored. We present Echo, a framework that combines task status signal learning without frame-level progress annotations with a compact design for status prediction and action generation within a single VLA. Specifically, Echo represents task status with stop and failure signals, added as two dimensions of each action chunk and denoised together with robot actions. Their combination distinguishes ongoing execution, successful completion, and failure. To learn these signals without fine-grained progress annotations, Echo constructs successful, failed, and recovery trajectories with complementary supervision properties. To enable status signal-guided action generation, Echo controls interactions between the signals and actions and denoises the signals faster than the actions. These allow later action denoising steps to condition on cleaner status estimates while preserving accurate signal prediction. Extensive real-world manipulation experiments show that Echo accurately predicts task status and consistently improves policy performance.

cs.RO↗

arXiv2026

MadStree: automated evaluation of tree-level de Sitter correlators

We first derived the formulae of recursive reduction and dlog-form differential equations for arbitrary tree-level cosmological correlator, taking the external relations of massless bulk propagators into consideration. The differential equations could be systematically solved and provide the efficient evaluation of them. We have implemented this methodology as a user-friendly Mathematica package MadStree for computing arbitrary tree-level cosmological correlators.

hep-ph↗

arXiv2026

Where RL Breaks: Localizing and Repairing Failures via Targeted Continued Pretraining

Reinforcement learning with verifiable rewards (RLVR) is widely used to improve reasoning in language models, yet growing evidence suggests that it mostly sharpens what the base model can already do. This raises a practical question: when RL fails, what is the base model missing? We look inside the RL policy's own rollouts and find that failures are localized: whether a rollout fails is strongly predicted by a small set of characteristic n-grams, ranging from reflection phrases such as "let's recheck" to specific operations such as "the inverse of A". These failures persist throughout RL training, although they are not beyond the model's reach: other sampled answers to the same question are often correct. Building on this observation, we use the failure $n$-grams to retrieve text from the pretraining corpus and continue pretraining the base model on it before RL. The resulting corpus has only 35M tokens, 3--4 orders of magnitude smaller than typical mid-training. We evaluate this approach on OLMo2 and Qwen3 models trained with RL on (1) GSM8K; (2) a mixture of math, commonsense, and knowledge questions; and (3) general math and science reasoning. Despite its size, the corpus yields consistent gains: our approach outperforms continued pretraining on the same amount of text sampled from the pretraining corpus across 10 out of 11 settings, including by $2.0$ points on GSM8K when inserted before mid-training. The retrieved text also transfers across model families: text retrieved for OLMo2-1B's failures improves Qwen3-4B on the mixture by $1.9$ points over the same baseline. Together, our results suggest that where RL fails is a cheap and effective signal for selecting pretraining data.

cs.CL↗

arXiv2026

Kalman-Bucy Filtering with Randomized Sensing: Fundamental Limits and Sensor Network Design for Field Estimation

Stability analysis of the Kalman filter under randomly lost measurements has been widely studied. We revisit this problem in a general continuous-time framework, where both the measurement matrix and noise covariance evolve as random processes, capturing variability in sensing locations. Within this setting, we derive a closed-form upper bound on the expected estimation covariance for continuous-time Kalman filtering. We then apply this framework to spatiotemporal field estimation, where the field is modeled as a Gaussian process observed by randomly located, noisy sensors. Using clarity, introduced in our earlier work as a rescaled form of the differential entropy of a random variable, we establish a lower bound on the expected spatially averaged clarity and show that its steady-state value converges to a grid-independent limit as the grid is refined. This result exposes fundamental performance limits through a composite sensing parameter that jointly captures the effects of the number of sensors, noise level, and measurement frequency. Simulations show close agreement between the continuous-time analytical solution and the discrete-time covariance, with the agreement improving as the sampling interval decreases. The analytical solution also avoids the recursive computations required in the discrete-time formulation. Most importantly, the derived limits provide principled and efficient guidelines for sensor network design prior to deployment.

eess.SY↗

arXiv2026

SEAR: Self-Evolving Agentic Robotics

Frontier language-model agents can now improve robots on their own: they write control code, test it, and keep what works as skills and memories for the next task. We introduce SEAR (Self-Evolving Agentic Robotics), a framework and benchmark built around two questions. Is the agent really learning, or just trying more often? And does what it learns still help on new tasks, in a new simulator, and on a real robot? The framework runs an agent in three modes: it writes code that becomes the policy (code-as-policy), acts live in the control loop (agent-as-policy), or works as a researcher that fine-tunes a vision-language-action model (auto-research). The benchmark adds controls that separate retained experience from repeated attempts, familiar tasks, privileged simulator state, larger budgets, and luck. Across seven frontier language models, 330 simulated and five physical tasks, and over four years of agent runtime, we highlight three findings. First, the learning curve of a robotic agent depends both on the overall time budget and the task difficulty relative to model capability. Harder tasks make longer sessions worthwhile; on easier tasks, re-rolling short attempts does just as well. Second, control code and shared routines transfer across tasks and simulators, while task-specific solutions do not. On a real robot, we observe a performance gap between agent-designed practice and novel human-designed scenes. Third, different models evolve more effectively in different ways. Some do best acting live, others writing code first. And a text-only model keeps pace with a comparable multimodal one: perception helps with error assessment, but agentic reasoning and tool use can matter more. We release the benchmark and codebase to support future research.

cs.RO↗

arXiv2026

LIBERO-RIG: A Hardware-Setup Robustness Benchmark for Vision-Language-Action Models

Robot policies are expected to generalize across deployment environments, including variations in the robot itself: cameras are bumped and remounted, calibrations drift, controllers run at different rates, and end-effector pads wear down. We present LIBERO-RIG, a benchmark that perturbs only the robot's hardware and setup while holding every LIBERO task fixed. The benchmark spans eleven perturbation axes, including camera mounting and optics, camera-to-arm calibration, actuation gain, control rate, proprioceptive noise, base pose, table height, and gripper pads. Each axis has five severity levels, yielding 210 conditions across the four LIBERO suites. We evaluate nine recent vision-language-action models using 500 episodes per condition and substantiate each finding with rigorous statistical tests. Pooled success falls from 0.932 to 0.767 at the middle severity and 0.509 at the highest severity. Performance degradation varies substantially across perturbation axes and action interfaces, and training on visual perturbations does not mitigate faults that leave camera images unchanged. We release the benchmark together with manifests specifying the fixed perturbation parameters for each episode.

cs.RO↗

arXiv2026

Learning from Own Solutions: Self-Conditioned Credit Assignment for Reinforcement Learning with Verifiable Rewards

Reinforcement learning with verifiable rewards (RLVR) has driven substantial progress in training LLMs for reasoning tasks, but representative methods such as GRPO assign uniform credit across all tokens, wasting gradient on routine tokens while under-crediting pivotal reasoning steps. Existing token-level credit assignment methods require resources beyond the model's own rollouts. GRPO variants rely on process reward models or ground-truth answers. Knowledge distillation assigns credit through per-token divergence but requires external teachers (On-Policy Distillation) or privileged information (On-Policy Self Distillation). However, these dependencies limit applicability in the pure RLVR setting. We observe that conditioning the model on its own verified trajectories induces a measurable per-token KL divergence between the original and conditioned distributions, and prove that distilling from a self-teacher constructed by verified trajectories leads to infeasible weighted-average solutions when multiple verified trajectories exist. We propose SC-GRPO (Self-Conditioned GRPO), which uses KL divergence mentioned before as a multiplicative weight on GRPO gradients. Across five benchmarks spanning math, code, and agentic tasks, SC-GRPO consistently outperforms 8.1% over GRPO and 5.9% over DAPO with stronger OOD performance. Moreover, SC-GRPO achieves higher performance than OPD.

cs.LG↗

arXiv2026

IR-SIM: A Lightweight Declarative Simulator for Navigation Learning and Benchmarking

Developing navigation policies requires simulation scenarios that support repeatable training and evaluation. Despite the availability of numerous simulators, constructing diverse navigation scenarios often involves writing simulator-specific code, using complex graphical interfaces, performing substantial manual configuration, or relying on resource-intensive computing platforms, making scenarios difficult to reproduce and reuse across experiments. To this end, we develop the Intelligent Robot Simulator (IR-SIM), a lightweight declarative simulator implemented as a Python library to support navigation learning and benchmarking through rapid construction of reusable and diverse scenarios and efficient execution on accessible platforms. IR-SIM represents scenarios as human-readable YAML configurations, in which necessary objects, behaviors, sensors, maps, and environment parameters can be flexibly specified and composed. This declarative representation makes IR-SIM friendly to large language model (LLM)-powered agents, enabling them to compose and modify executable navigation scenario files through the provided agent skills, instead of writing a large amount of code that is difficult to reuse. Despite its lightweight implementation, IR-SIM provides the essential components for navigation simulation, while retaining interfaces to high-fidelity simulators for downstream validation. Experiments demonstrate that IR-SIM runs up to several hundred times faster on the evaluated CPU platform, while agent skills reduce mean LLM-based scenario construction time by more than $50\%$ across both evaluated models. The experiments further show ability to support reproducible benchmarking and RL-based navigation policy learning.

cs.RO↗

arXiv2026

Joint DNN Partitioning and Resource Allocation for Satellite-Terrestrial Collaborative Inference Systems

This letter investigates a satellite-terrestrial collaborative inference system for remote sensing information processing, where a ground station (GS) serves multiple low Earth orbit (LEO) satellites running heterogeneous deep neural network (DNN) inference tasks. By partitioning the DNN between the satellite and the GS, each satellite executes the front-end layers locally, transmits the resulting intermediate features to the GS, thereby offloading the remaining layers to the GS for inference completion. We jointly optimize DNN partitioning, satellite and GS computing resources, satellite-ground bandwidth allocation, and satellite transmit power to minimize the average task completion latency under per-satellite energy constraints. To address the coupling between discrete partitioning decisions and continuous communication-computing variables, we propose a two-layer optimization algorithm. The inner layer performs closed-form updates and efficient nested bisection searching to update the continuous resource variables, while the outer layer employs random-restart coordinate descent to refine the partitioning decisions. Simulation results show that the proposed algorithm effectively reduces task completion latency compared with conventional binary offloading.

eess.SP↗

arXiv2026

Intelligent Beamforming and Handover via Physics-Informed Beam-Aware CKM Diffusion

Downlink communications from base stations (BSs) to unmanned aerial vehicles (UAVs) in sixth-generation (6G) networks require precise beam alignment to overcome severe mobile communication path loss. However, traditional exhaustive beam sweeping relies on discrete codebooks and consumes valuable air-interface resources for online measurements, rendering it inefficient for highly dynamic aerial environments. In this paper, we propose BeamCKMDiff, a physics-informed generative diffusion framework designed to construct high-fidelity continuous beam-aware channel knowledge maps (BeamCKMs). Unlike existing empirical approaches, BeamCKMDiff employs a diffusion transformer (DiT) architecture featuring a dual-path conditioning mechanism. It integrates an analytical line-of-sight (LoS) beam prior with environmental topologies, while simultaneously embedding continuous beamforming vectors through an adaptive layer normalization (adaLN) mechanism. Building upon this differentiable generative mapping, we propose a CKM-based end-to-end continuous beamforming and proactive dual-BS handover algorithm. By analytically propagating gradients through the reverse diffusion process, the proposed framework enables successive convex approximation (SCA)-based beamforming optimization in the computational domain, effectively bypassing physical pilot scanning. Simulation results demonstrate that BeamCKMDiff achieves a normalized mean square error (NMSE) of --23.96 dB, establishing a highly reliable spatial prior. Compared to discrete beam sweeping and LoS-assumed baselines, the proposed CKM-based beamforming adapts to non-LoS (NLoS) conditions, providing accurate beam alignment and higher spectral efficiency. Furthermore, the integrated dual-BS handover mechanism eliminates blockage-induced link outages, ensuring robust connectivity for high-mobility aerial platforms.

eess.SP↗

arXiv2026

DiffCVaR: Reinforcement Learning for Risk Adaptation via Differentiable CVaR Barrier Functions

Planning through crowded environments under uncertain obstacle motions remains difficult, as stochastic interactions often induce overly conservative behavior or reduced efficiency. To address this challenge, we propose an end-to-end risk adaptation framework for crowd navigation under obstacle-motion uncertainty modeled by a Gaussian mixture model. The framework combines reinforcement learning~(RL) with a differentiable quadratic-program safety layer based on Conditional Value-at-Risk~(CVaR) barrier functions, jointly learning nominal control input, risk level, and safety margin and enforcing explicit probabilistic safety constraints. This design enables context-aware adaptation, promoting efficient behavior while invoking caution only when necessary. We conduct extensive evaluations in dynamic, uncertain, and crowded environments across varying obstacle densities and robot models, and further assess generalization under three out-of-distribution cases. Comparisons across optimization-based, RL-based, and integrated RL and optimization methods are provided, and the proposed method is shown to deliver the strongest overall performance in safety, efficiency, and generalization under uncertainty.

cs.RO↗

arXiv2026

ReVeal: A Reconstruction-Aware Real-to-Sim Framework for VLA Policy Evaluation

Simulation-based evaluation provides a scalable and repeatable alternative to real-world evaluation of vision-language-action (VLA) policies. However, reconstruction errors can cause simulated policy performance to diverge from real-world performance, motivating the need to assess reconstructed environments for downstream VLA policy evaluation. We present ReVeal, a real-to-sim assessment framework combining workspace reconstruction, reconstruction-level assessment, and matched closed-loop policy evaluation. Novel-View Mesh Fidelity (NVMF) and Annotated Planar Geometry Fidelity (APGF) assess observation and planar geometric fidelity, respectively. We also develop PGSR-D, a reconstruction pipeline incorporating monocular depth supervision to improve geometry where multi-view visual cues are limited. Across 8 assessment scenes, NVMF and APGF consistently distinguish the fidelity of 2DGS, PGSR, and PGSR-D. Matched evaluations of GR00T, SmolVLA, and pi0.5 across 8 humanoid manipulation tasks show consistent ordering between reconstruction fidelity and real-sim performance agreement across pipelines. Further analysis of the evaluation workspaces shows that higher fidelity is associated with stronger real-sim agreement.

cs.RO↗

arXiv2026

Distributed Model-Based Diffusion For Scalable Multi-Robot Trajectory Optimization

Trajectory optimization for multi-robot systems remains a critical challenge, particularly when navigating highly non-convex, non-linear, and non-differentiable environments. While Model-Based Diffusion (MBD) has recently emerged as a promising sampling-based optimization paradigm for single-robot trajectory generation, extending it to multi-robot systems results in a centralized, high-dimensional inference problem that (i) suffers from poor sample efficiency due to the curse of dimensionality and (ii) requires global access to all robots' dynamics, constraints, and objectives. To address this, we propose Distributed Model-Based Diffusion (DMBD), a distributed server-robot method that decomposes the reverse diffusion process into local conditional reverse diffusion processes. This decomposition enables each robot to iteratively perform denoising independently within its own control subspace while conditioning on the current trajectory estimates of the other robots that are aggregated and broadcast by the server. Extensive simulations in goal swapping, multi-floor coverage, parking, and rush-hour scenarios demonstrate that DMBD achieves strong scalability, solving many challenging coordination tasks with sub-second computation time and outperforming existing baselines.

cs.RO↗

arXiv2026

Probabilistic Symbol-Level Precoding based Affine Frequency Division Multiplexing Transmission

Affine frequency division multiplexing (AFDM) has recently gained significant attention due to its robustness against time-frequency doubly selective channel fading. However, the high computational complexity at the receiver poses a critical challenge for practical deployment. To overcome this issue, we propose a probabilistic symbol-level precoding (SLP)-based AFDM transmission framework, in which the processing burden in downlink transmission is shifted from the user to the base station (BS), enabling direct symbol detection without channel estimation or equalization at the receiver. In the proposed framework, the BS exploits the uplink channel state information (CSI) to design the downlink transmit waveform based on uplink-downlink channel reciprocity. In particular, we innovatively introduce a probabilistic SLP technology by explicitly characterizing the likelihood of symbol detection errors under noise perturbations. Specifically, the transmitted symbols are optimized to minimize the likelihood that the received symbols fall into erroneous decision regions, where the resulting error-probability minimization problem is subsequently approximated as a second order cone programming (SOCP) problem by exploiting the monotonicity of the objective function. Simulation results show that the proposed probabilistic SLP-based scheme achieves performance comparable to that of conventional AFDM receivers, whilst enjoying significant reduction of computational complexity at the receiver end. These results demonstrate the effectiveness and practical potential of the proposed approach.

eess.SP↗

arXiv2026

Standardizing Longitudinal Radiology Report Evaluation via Large Language Model Annotation

Longitudinal information in radiology reports refers to the sequential tracking of findings across multiple examinations over time, which is crucial for monitoring disease progression and guiding clinical decisions. Many recent automated radiology report generation methods are designed to capture longitudinal information; however, validating their performance is challenging. There is no proper tool to consistently label temporal changes in both ground-truth and model-generated texts for meaningful comparisons. Large language models (LLMs) offer a promising annotation alternative, as they are capable of capturing nuanced linguistic patterns and semantic similarities without extensive manual intervention. They also adapt well to new contexts. In this study, we therefore propose an LLM-based pipeline to automatically annotate longitudinal information in radiology reports. The pipeline first identifies sentences containing relevant information and then extracts the progression of diseases. We evaluate and compare five mainstream LLMs on these two tasks using 500 manually annotated reports. Considering both efficiency and performance, Qwen2.5-32B was subsequently selected and used to annotate another 95,169 reports from the public MIMIC-CXR dataset. Our Qwen2.5-32B-annotated dataset provided us with a standardized benchmark for evaluating report generation models. Using this new benchmark, we assessed seven state-of-the-art report generation models. Our LLM-based annotation method outperforms existing annotation solutions, achieving 11.3\% and 5.3\% higher F1-scores for longitudinal information detection and disease tracking, respectively. The source code is available at https://github.com/wxinyi1996/Standardizing-Longitudinal-Chest-X-ray-Report-Evaluation-via-Large-Language-Model-Annotation.git.

cs.CL↗

arXiv2026

SkillForge: Compositional Skill Synthesis with Verification-in-the-Loop for Generating Formally Verified Dafny Programs

Generating formally verified programs from natural language remains challenging: existing approaches either produce code in a single pass without recourse when verification fails, or rely on open-ended agentic reasoning that is non-deterministic and opaque. We introduce SKILLFORGE, a framework that decomposes formal code synthesis into a library of atomic, reusable skills, each targeting a specific subtask such as specification inference, body synthesis, invariant generation, error diagnosis, or targeted repair, and defined by a prompt template, tool binding, and decidable success criterion. A verification-driven harness orchestrates these skills: it submits candidates to the Dafny verifier, diagnoses failures into structured categories, deterministically routes to the appropriate repair skill, and iterates until formal correctness is proved or a budget is exhausted. On a curated benchmark of natural language to Dafny specification pairs, SKILLFORGE substantially outperforms both state-of-the-art agentic approaches (including ReAct-style agents, MCTS-based repair, and RL-guided verification) and traditional iterative baselines, while requiring fewer tokens and lower latency. Ablation studies confirm that every skill contributes measurably, and the harness converges rapidly with the majority of programs verified on the first attempt.

cs.CL↗