Searcharxiv⌕ Search

arXiv subjects

Xuanyu Liu

Publications and source records attributed to Xuanyu Liu.

At least 19 recordsLinked to original sources

WiFo-R: Wireless Foundation Model with Recurrent Channel Representation

Wireless foundation models learn transferable channel state information (CSI) representations across heterogeneous wireless systems. However, conventional grid-aligned representations couple feature sequence length to CSI dimensions, restricting the allocation of representation capacity to different propagation conditions and task requirements while increasing inference overhead. This paper proposes WiFo-R, a wireless foundation model that learns progressive channel representations with selectable capacity. A channel-query cross-attention encoder aggregates spatial, temporal, and spectral CSI information, while parameter-shared recurrent updates refine existing channel features and introduce additional latent tokens. Multi-mask, multi-depth self-supervised pretraining equips intermediate states with strong reconstruction capabilities, followed by task-aligned post-training for channel estimation and CSI feedback. WiFo-R supports temporal and frequency-domain channel prediction, channel estimation, and variable-rate CSI feedback, and transfers to localization and LoS/NLoS classification. Experiments across in-distribution and out-of-distribution channel configurations demonstrate accurate reconstruction, progressive performance improvements, and strong transferability. A single feedback encoder--decoder supports six payload lengths at fixed quantization precision, while cross-attention and selectable recurrent depth improve the accuracy--complexity trade-off for large-scale CSI observations.

eess.SP↗

What Makes Good Datasets for Intelligent Multi-Modal Sensing-Communication Integration?

Intelligent multi-modal sensing-communication integration, namely Synesthesia of Machines (SoM), represents an AI-native paradigm for enabling embodied intelligence and future wireless networks. Multi-modal sensing and communication foundation models developed for SoM learn cross-modal associations and generalizable environmental representations through large-scale pretraining, making the quality of pretraining datasets critical. However, existing methods lack unified evaluation of heterogeneous multi-modal data and thus struggle to meet the requirements for multi-modal pretraining dataset construction and selection. To bridge this gap, we introduce a comprehensive data evaluation framework for intelligent multi-modal sensing-communication integration. Specifically, we establish two complementary heterogeneous multi-modal data evaluation dimensions, including intrinsic data quality and validity, and model-based data utility. Furthermore, we present case studies to validate the effectiveness of the two evaluation dimensions. We also extend our analysis to investigate the impact of data quality on pretraining scaling laws and explore the value of integrating measurement and simulation data. Finally, we outline promising directions for future research on data evaluation.

eess.SP↗

LiveGeo: Benchmarking Geographic Agents in Continuously Evolving Worlds

We introduce GeoWorld, a controllable and continuously evolving geographic environment platform for LLM agents, and LiveGeo, a benchmark for evaluating reliable decision-making in real-world geographic environments using GeoWorld. Unlike existing evaluations that largely reason and plan over fixed geographic states, GeoWorld is grounded in real-world geographic structure and allows external world states to evolve along their own temporal processes while preserving verifiable state histories. LiveGeo places geographic tasks within this evolving environment, requiring agents to actively acquire task-relevant information, verify whether the evidence supporting prior decisions remains valid, and revise their plans when necessary. Our results reveal that even frontier LLMs struggle with static tasks, with no model consistently excelling across tasks, while strong static performance does not ensure dynamic reliability. Further analysis identifies a key bottleneck: agents struggle to acquire and translate relevant evidence into valid decisions and maintain their correctness as the world evolves. By releasing LiveGeo together with its GeoWorld platform, we aim to provide the community with a flexible and extensible infrastructure for studying the next generation of agents capable of reliable action in evolving geographic worlds.

cs.AI↗

VLA-Rolling: Fewer Denoising Steps, Better Control via Adaptive Action Reuse

Flow-matching vision-language-action (VLA) models generate an action chunk by iteratively denoising Gaussian noise, but typically execute only a short prefix before replanning with updated observations. Under this receding-horizon paradigm, the unexecuted remainder---the action tail---is discarded, and the next chunk is regenerated from fresh noise. We show that this discarded tail retains transferable denoising progress: action-tail initialization can match or outperform a native 10-step Gaussian baseline using fewer denoising steps. However, the optimal denoising budget varies substantially across samples, indicating that tail reuse should be adapted to the updated environment. Based on this observation, we introduce VLA-Rolling, a lightweight inference framework that reuses action tails through three stages: action-tail inheritance, per-action confidence estimation, and confidence-guided refinement through renoising and denoising budget allocation, while the underlying VLA remains frozen. Across $π_{0.5}$, SmolVLA, X-VLA, and MolmoAct2 on LIBERO, MetaWorld, LIBERO-Plus, and RoboTwin 2.0, VLA-Rolling reduces action-chunk inference cost in all 16 model--benchmark pairs, achieving up to a $1.48\times$ speedup, while improving success rate in 14 pairs by up to $15.08$ percentage points. These results show that action tails are not merely discarded predictions, but reusable intermediate states that enable more efficient and accurate VLA inference. These results demonstrate that the action tail can support the generation of better actions with fewer denoising steps. Our project is available at https://chenbao123ww.github.io/VLA-Rolling/.

cs.AI↗

Scene-Designer: Learning Design Intelligence for Agentic 3D Scene Generation

Generating complete, coherent, and usable 3D indoor scenes from natural language requires strong \textbf{design intelligence}: the ability to design high-quality scenes and critique their quality. However, existing approaches either rely on general-purpose models lacking domain-specific design experience or learn from static scenes. In this work, we introduce \textbf{Scene-Designer}, an agentic framework that learns design intelligence through complementary design and critique experiences. We construct two datasets: \textbf{Scene-OTA-35K}, which provides Observation--Thought--Action supervision for learning scene design and efficient error correction, and \textbf{Scene-OTC-15K}, which provides Observation--Thought--Critic supervision for learning scene quality critique across physical validity, functional correctness, and layout coherence. Based on these experiences, we propose \textbf{Scene-GRPO}, a Critic-guided GRPO framework that improves design decisions and reduces unnecessary trial-and-error corrections. Extensive experiments demonstrate that Scene-Designer achieves superior performance in physical validity, functional correctness, and layout coherence, while the learned Scene-Critic significantly outperforms general-purpose models in scene critique. Moreover, test-time Critic guidance further improves scene generation, showing that the Critic can provide useful guidance beyond training. Additional analysis suggests a trend of mutual enhancement between scene design and critique abilities.

cs.CV↗

FreeBrush: Motor Signals-Based Toothbrushing Monitoring System for Electric Toothbrushes

Oral health is vital for overall well-being, yet many users brush different tooth surfaces for insufficient or excessive durations. Toothbrushing monitoring systems can help improve brushing habits by reminding users of under-brushed and over-brushed surfaces, but existing approaches often rely on external sensors, increasing cost and reducing practicality. We observe that the intrinsic motor signals of electric toothbrushes can reflect the brushing surfaces and durations. When brushing different tooth surfaces, the brush head experiences different constraints from the surrounding oral cavity, changing the motor's driving currents and position feedback. Based on this observation, we propose FreeBrush, a toothbrushing monitoring system that uses the intrinsic motor signals of electric toothbrushes to recognize brushed tooth surfaces and estimate brushing durations without any external sensors. We prototype the system on a commercial off-the-shelf electric toothbrush and conduct a comprehensive evaluation with 18 participants. FreeBrush achieves 92.8% accuracy for 16-surface recognition and 92.0% accuracy for brushing duration estimation. The system further shows robustness across different toothpastes, brush heads, vibration modes, usage scenarios, and long-term use, and is capable of running on edge devices. By recognizing brushed tooth surfaces and estimating the brushing duration, FreeBrush can help users identify under-brushed and over-brushed surfaces and build balanced brushing habits.

cs.HC↗

Learning Robust Recommenders from Noisy Implicit Feedback via GMM-Weighted Bayesian Transition Matrix

Label noise is a central challenge in learning from implicit feedback for recommendation. Conventional approaches discard noisy examples for robustness, but this sacrifices data efficiency. Unlike filtering approaches, Bayes-label transition matrix (BLTM) based methods keep all data, but their transition matrix estimates are skewed in practice. To reduce this skew, we introduce GMM-weighted Bayes-label Transition Matrix (RGBT), which augments BLTM with GMM-based instance weights. A GMM assigns each instance a reliability score, and these scores calibrate the BLTM to reduce bias. We show theoretically that RGBT retains all samples, yields consistent estimates, and provably reduces variance compared to CLTM, generalizing to any method with deterministic labels and non-negative weights. Experiments on real and synthetic datasets show that RGBT handles noisy samples more effectively than sample-selection methods, and calibrates the transition matrix more accurately than existing approaches.

cs.LG↗

EmbodiedMemory-Bench: Benchmarking Embodied Memory for Long-Horizon Embodied Tasks

Long-horizon embodied interaction requires agents to retain and continually update information about the environment as they observe, act, and encounter change. Yet current agents struggle to maintain such memory reliably. Our analysis traces this limitation to four key deficiencies: weak fine-grained visual memory, unreliable dynamic world-state tracking, failing to record world state revealed by interaction outcomes, and limited generalization from prior experience. However, existing benchmarks do not directly assess these memory capabilities during long-horizon embodied interaction. To address this gap, we introduce EmbodiedMemory-Bench (EMem-Bench), comprising 2,554 interactive episodes across four task families. EMem-Bench requires agents to build and update memory from interaction history, then use it to complete a later task by acting in the environment. We further present Embodied-Memorizer (EMem), an external memory system that organizes embodied experience into spatial, event, and scene memories. We also train EMem-8B, an 8B policy that manages and uses these memories. We evaluate a diverse range of open-source and proprietary MLLMs and representative multimodal memory systems. Results show that current models remain weak and uneven across the four challenges. Under matched backbones, EMem achieves the best overall performance among the evaluated memory systems and improves both open-source and proprietary models, while EMem-8B further improves over its backbone. Project page: https://zju-omniai.github.io/EmbodiedMemoryBench/

cs.CV↗

ReCalMatch:Reliability-Calibrated Semantic Guidance for Semi-Supervised Fine-Grained Recognition

Semi-supervised fine-grained visual recognition is highly vulnerable to overconfident pseudo-label errors: visually similar categories frequently produce high-confidence yet incorrect predictions, and consistency regularization then reinforces these errors throughout training. Existing semi-supervised learning (SSL) methods estimate pseudo-label reliability almost entirely from the visual classifier itself---maximum probability, adaptive thresholds, or entropy---signals that remain blind to whether a predicted class is \emph{semantically} compatible with the visual representation. We propose \textbf{ReCalMatch}, a reliability-calibrated semantic framework for semi-supervised fine-grained recognition. Rather than treating textual semantics as auxiliary supervision, ReCalMatch uses multi-aspect semantic prototypes as \emph{calibration evidence} for pseudo-label learning. We construct class-conditioned semantic prototypes from class names and domain-specific semantic aspects, and measure a \emph{visual--semantic agreement} score between each unlabeled embedding and its pseudo-label prototype. This agreement is combined with prediction confidence and entropy into a single reliability weight that down-weights pseudo-labels that are visually confident but semantically inconsistent. A semantic consistency term and a semantic margin regularizer further sharpen prototype separability under limited labels. Extensive experiments on CUB-200-2011, Stanford Dogs, NABirds, and iNaturalist18 show that ReCalMatch consistently improves strong SSL baselines, with the largest gains in low-label regimes where pseudo-label noise is most severe.

cs.CV↗

Embodied-Navigator: Point, Think, Memorize, and Align for Efficient Navigation

Although Large Vision-Language Models (VLMs) have significantly advanced embodied navigation, their direct deployment remains challenging, as existing methods often force VLMs into unnatural action spaces that misalign with their 2D pre-training priors, compounded by rigid reasoning schedules and inefficient memory management. To overcome these limitations, we propose TAMP-Nav, a unified framework for efficient embodied navigation. First, we introduce a Pixel-to-3D Action Formulation (Point) that reformulates navigation into 2D visual prompting. Specifically, the VLM merely selects 2D pixels, which are then projected into 3D coordinates for a low-level SLAM controller. This design naturally aligns embodied execution with the VLM's inherent 2D visual capabilities. Second, we propose an integrated Selective Reasoning and Anchor-Trajectory Memory mechanism (Think and Memorize), which dynamically triggers Chain-of-Thought and retains high-fidelity memory only at critical nodes, compressing redundant trajectories into lightweight Space-Time Indicators, thereby preserving critical historical information and enhancing spatio-temporal perception. Finally, we design an efficient Two-Level Alignment Paradigm (Align) via Group Relative Policy Optimization (GRPO). By superimposing global outcome rewards with fine-grained process rewards, this dense supervision tightly aligns the agent's cognitive planning with physical environmental feedback, endowing the model with adaptive reasoning capabilities. Experiments demonstrate that TAMP-Nav achieves state-of-the-art performance (e.g., 66.2% SR on R2R-CE) with high runtime and sample efficiency (requiring only 90k training trajectories).

cs.RO↗

Interference-engineered shortcut to perfect state transfer

Achieving fast, high-fidelity state transfer is fundamental to scalable integrated photonics and quantum information processing. While adiabatic evolution provides inherent robustness against control and fabrication imperfections, its requirement for slow driving leads to impractically long propagation distances in photonic circuits. Existing acceleration strategies, such as shortcuts to adiabaticity (STA), can dramatically shorten evolution times but generally rely on non-native auxiliary couplings or delicate Hamiltonian engineering that are difficult to implement in practice. Here we introduce evolution-pause synthesis (EPS), an interference engineered shortcut protocol that achieves fast, near-perfect state transfer strictly within the native system Hamiltonian. It achieves this by treating transient excitations as coherent resources and canceling their accumulated amplitudes via strategically interleaved pauses. By decoupling relative dynamical phase accumulation from parameter variations, EPS steers open transition trajectories into a closed loop in complex amplitude space, enabling perfect state transfer without auxiliary fields or complex parameter detours. We demonstrate this mechanism in Landau-Zener dynamics and extend it to a multilevel STIRAP process, achieving an 11.8-fold acceleration over the adiabatic baseline. Further, we experimentally validate EPS on a silicon photonic platform, realizing high-fidelity state transfer in a $16\,μ\mathrm{m}$ footprint, a nearly tenfold reduction in device length compared with a $150\,μ\mathrm{m}$ adiabatic reference. EPS offers a general hardware-compatible framework for fast, practical coherent control across wave and quantum platforms.

physics.optics↗

WiFo-MiSAC: A Wireless Foundation Model for Multimodal Sensing and Communication Integration via Synesthesia of Machines (SoM)

Current learning-based wireless methods struggle with generalization due to the fragmented processing of communication and sensing data. WiFo-MiSAC addresses this as a task-agnostic foundation model that tokenizes heterogeneous signals into a unified space for self-supervised pre-training. A shared-specific disentangled mixture-of-experts (SS-DMoE) architecture is employed to decouple modality-shared and modality-specific representations, facilitating interaction without cross-modal interference. By combining masked reconstruction with contrastive alignment, the model achieves state-of-the-art performance across downstream tasks, including beam prediction and channel estimation. Experimental results demonstrate robust few-shot adaptation and seamless integration of new modalities, positioning WiFo-MiSAC as a scalable backbone for future integrated sensing and communication systems.

eess.SP↗

Triple Expert Learning from Noisy Labels for Semi-Supervised Vision Foundation Model Adaptation

Semi-supervised adaptation of vision foundation models (VFMs) commonly freezes the pretrained backbone and updates lightweight modules such as LoRA. However, pseudo-labels have mixed reliability, and a single LoRA adapter must absorb reliable, ambiguous, and noisy gradients in the same low-rank space. This can make VFM adaptation sensitive to pseudo-label noise. We propose \textbf{TriNoL}, a \textbf{Tri}ple-expert learning framework from \textbf{No}isy \textbf{L}abels for semi-supervised VFM adaptation. TriNoL routes unlabeled samples into three confidence regions and assigns them to three LoRA experts: a Positive Expert for high-confidence pseudo-labels, an Alignment Expert for medium-confidence ambiguous samples, and a Negative Expert for low-confidence noisy samples. The VFM backbone remains frozen, and only the LoRA experts and classifier head are updated. By separating different pseudo-label reliability regions into specialized adaptation paths, TriNoL improves robustness to noisy supervision while keeping the training cost low.

cs.CV↗

WiFo-INR: A Wireless Foundation Model Based on Implicit Neural Representations

Wireless foundation models are emerging as a promising paradigm for AI-native physical-layer design. However, existing methods typically model channel state information (CSI) as image-like discrete tensors with generic token decoders that may struggle to capture complex high-frequency variations efficiently and often produce high-dimensional, size-dependent representations. In this paper, we propose WiFo-INR, an implicit neural representation (INR)-based wireless foundation model that represents CSI as a coordinate-conditioned neural function. A Transformer encoder maps partial or coarse CSI to fixed-dimensional modulation tokens that adapt a SIREN-based decoder, and a compression autoencoder enables quantized CSI feedback. It adopts a two-stage self-supervised pretraining scheme, where mixed masking and denoising improve channel reconstruction and compression-enhanced pretraining enables accurate CSI feedback at low compression ratios. Extensive experiments demonstrate that WiFo-INR learns efficient, compact, and CSI-size-independent implicit wireless representations. Compared with existing foundation models, WiFo-INR improves channel reconstruction and CSI feedback performance while substantially reducing inference latency. It also transfers efficiently to diverse wireless tasks with minimal fine-tuning overhead and achieves zero-shot generalization to unseen CSI sizes.

eess.SP↗

The normalized expectation-maximization (N-EM) algorithm

Although the $\textit{expectation-maximization}$ (EM) algorithm is a powerful optimization tool in statistics, it can only be applied to missing/incomplete data problems or to problems with a latent-variable structure. It is well known that the introduction of latent variables (or the data augmentation) is an art; i.e., it could only be done case by case. In this paper, we propose a new algorithm, a so-called $\textit{normalized EM}$ (N-EM) algorithm, for a class of log-likelihood functions with integrals. As an extension of the original EM algorithm, the N-EM algorithm inherits all advantages of EM-type algorithms and consists of three steps: normalization step (N-step), expectation step (E-step) and maximization step (M-step), where the N-step is to construct a $\textit{normalized density function}$ (ndf), the E-step is to compute a well-established surrogate $Q$-function and the M-step is to maximize the $Q$-function as in the original EM algorithm. The ascent property, the best choice of the ndf, and those N-EM algorithms with a difficult M-step are also explored. By multiple real applications, we have shown that the N-EM algorithm can solve some problems which cannot be addressed by the EM algorithm. Next, for problems to which the EM can be applied (often case by case), the N-EM algorithm can be employed in a unified framework. Numerical experiments are performed and convergence properties are also established.

stat.ME↗

WiFo-2: a generalist foundation model unifies heterogeneous wireless system design

Emerging sixth-generation wireless systems are increasingly heterogeneous, with compatibility across diverse configurations, ubiquitous coverage, and expanded functionalities. Although deep learning has substantially benefited wireless system design, existing approaches are typically trained for specific system settings and scenarios with limited generalizability. Here we present WiFo-2, a space-time-frequency foundation model for unified wireless communications and sensing system design. Pretrained on a heterogeneous dataset of 11.6 billion channel state information (CSI) points, WiFo-2 learns generalized wireless representations across scenarios, configurations, and tasks, and exhibits scaling-law behavior. WiFo-2 achieves reliable and accurate zero-shot channel reconstruction, outperforming fully supervised task-specific models. With only 1% of the training samples required by supervised AI models, WiFo-2 achieves state-of-the-art performance across 9 distinct wireless tasks. A functional hardware prototype further demonstrates its real-world deployability and superior capability across diverse wireless tasks. This work provides a versatile wireless design framework and advances understanding of wireless channels.

eess.SP↗

Show, Don't Tell: Evaluating Spatial Cognition in Generative Pixels Rather Than LLM Text

Spatial intelligence is essential for agents to move from static semantic understanding toward interacting with the physical world. Many spatial tasks are grounded in continuous visual scenes, where locations, regions, and paths are more naturally expressed by pointing, marking, or drawing than by reporting precise coordinates or discrete textual symbols. Yet existing spatial reasoning benchmarks usually require coordinates, options, or text, creating an answer-interface mismatch for image-generation models. This makes it difficult to evaluate image-generation models under the same task semantics as text-output VLMs, despite their ability to externalize spatial judgments directly in pixel space. We propose ProVisE (Protocolized Visual Evaluation), a benchmark-agnostic framework that elicits protocol-constrained visual answers from image-generation models and parses them into structured predictions compatible with original metrics. ProVisE also includes an Agentic builder that constructs and validates task-specific protocols for new benchmarks. We further introduce SpatialGen-Bench, a curated diagnostic benchmark of 470 samples across 14 spatial subtasks, four capability levels, and diverse answer forms. We evaluate representative text-output VLMs and image-generation models in a unified setting and validate Agentic protocol construction on six external spatial benchmarks. Results show that image-generation models are competitive when spatial answers can be externalized directly in pixel space, while text-output VLMs retain a clear advantage in compositional spatial reasoning. These findings reveal complementary strengths of pixel-space expression and text-based reasoning and establish a metric-compatible testbed for studying spatial cognition in image-generation models.

cs.CV↗

HoloGeo: Mitigating Landmark Bias in Geo-localization via Evidence-Driven Reasoning

Recent advances in Vision-Language Models (VLMs) have significantly improved image geo-localization, yet existing models remain susceptible to landmark bias, causing them to overlook geographical cues or form spurious correlations, ultimately resulting in inaccurate localization. To systematically investigate this issue, we first design two quantitative metrics, Bias Intensity (BI) and Bias Harmfulness (BH), to characterize the impact of landmarks exerted on model reasoning, and establish a comprehensive benchmark, LandmarkBias-3K. To mitigate landmark bias, we further propose an evidence-driven reasoning framework, HoloGeo, to improve the reliability of geo-localization. HoloGeo is supported by a high-quality dataset, BF-30k, annotated with structured multi-evidence bias-free reasoning chains. By incorporating multi-dimensional rewards, HoloGeo explicitly encourages balanced attention over diverse visual cues and achieves evidence-driven joint reasoning. Extensive experiments demonstrate that HoloGeo not only maintains excellent performance on IM2GPS3K and YFCC4k but also significantly outperforms existing open-source VLMs on LandmarkBias-3K, validating its effectiveness for robust geospatial reasoning.

cs.CV↗