arXiv subjects

Peizheng Li

Publications and source records attributed to Peizheng Li.

At least 19 recordsLinked to original sources

arXiv2026

Beyond Object Selection:Markerless Gaze-based Robot Placement at Arbitrary Positions

Gaze-based assistive manipulation typically supports object selection, while arbitrary-position placement requires accurate spatial alignment between the headset and robot. However, for gaze-based manipulation, pose accuracy does not necessarily translate into task accuracy: translational and rotational errors jointly affect the transformed gaze ray and may compensate for each other. To study cross-device alignment from this task-oriented perspective, we present a markerless interaction framework and a dedicated cross-device dataset. We propose Graph-based Reference Selection to address sparse robot references. We further develop and benchmark multiple task-specific alignment pipelines under a unified protocol. Specifically, we introduce Gaze--Surface Intersection Error (GSIE), which directly measures the spatial error of the gaze-specified target. Experiments show that alignment methods ranked highly by conventional pose metrics are not always optimal in GSIE, demonstrating the importance of evaluating gaze-based manipulation at the task level.

cs.RO↗

arXiv2026

RegionCache: Semantic-Aware Region Reuse for Efficient Multi-Turn Image Generation

Real-world image generation often involves multi-turn editing, where users iteratively modify small regions while most image content remains unchanged. However, existing diffusion transformer (DiT)-based editing pipelines recompute the entire image at every turn, causing substantial redundant computation. Existing DiT acceleration methods further ignore semantic correspondence across prompts, leading to unnecessary recomputation or unsafe reuse that harms editing quality. To address this, we propose RegionCache, a semantic-aware reuse framework for multi-turn image editing that selectively reuses diffusion states from unchanged regions. RegionCache detects reusable regions through semantic overlap between consecutive prompts and cross-attention localization, and adopts an adaptive reuse schedule based on prompt similarity and contextual consistency. Experiments on PixArt-alpha demonstrate that RegionCache achieves 1.43x--2.55x end-to-end speedup while maintaining comparable image quality. Code is available at https://github.com/hebutBryant/RegionCache.

cs.CV↗

arXiv2026

TQD-Track: Temporal Query Denoising for 3D Multi-Object Tracking

Query denoising has become a standard training strategy for DETR-based detectors. Denoising queries, initialized by perturbing ground truths, share similarities with track queries in a typical DETR-based Multi-Object Tracking (MOT) method, warranting exploration of their potential synergy. However, query denoising in existing MOT methods is performed only within a single frame, preventing trackers from learning inter-frame temporal association from the denoising process. To address this issue, we propose TQD-Track, a Temporal Query Denoising (TQD) method tailored for MOT. In our method, denoising queries are initialized from ground truths in the previous frame and then propagated into the current frame in the same way as track queries, serving as additional independent data association candidates. These denoising queries carry temporal information and instance-specific feature representations, effectively emulating and augmenting track queries. Moreover, to simulate various real-world MOT challenges for robust tracking, we introduce several corresponding noise types to generate diverse denoising queries. We analyze the impact of our temporal query denoising for two tracking paradigms, tracking-by-attention and alternating detection and association, demonstrating its generalization. Extensive experiments on the nuScenes and Argoverse~2 datasets demonstrate that our approach consistently enhances different MOT baselines, requiring only modifications in the training process. Code and models are available at https://github.com/yutongy98/TQD-Track.

cs.CV↗

arXiv2026

When Robots Exchange Meaning: A Demo of Goal-Oriented Semantic Communications for Collaborative Robotics

Collaborative robotics is a representative task-oriented 6G use-case, where communication quality should be reflected in mission execution, environment understanding, and closed-loop operation rather than packet delivery alone. This demo paper presents a robot-edge semantic communication (SemCom) testbed integrating robot-side visual compression, edge-side semantic mapping, and dashboard-based mission interaction. A mobile robot equipped with RGB-D sensing and LiDAR runs ROS 2, while a Jetson Orin edge node performs reconstruction, RTAB-Map mapping, semantic object handling, and browserbased visualization. As an initial proof of concept, RGB frames are encoded on the robot into VQ-VAE tokens using an ONNX Runtime encoder and reconstructed on the edge using a PyTorch decoder. A 320 X 240 image is represented by an 80 X 60 token grid with a packed payload of 5400 bytes, corresponding to a 42.67X reduction relative to model-input RGB bytes. The reconstructed visual stream is further associated with depth, pose, and 3D mapping information to generate a semantic map for downstream robotic applications. The demo exposes the full path from semantic visual transport to object-level map interaction, and provides a practical platform for future task-aware 6G networking studies at the intersection of SemCom, embodied AI, and physical AI-enabled robotics. A video of the demo is available at https://tinyurl.com/Tos09

cs.RO↗

arXiv2026

The Price of Meaning: Quantifying Semantic Communication Overheads in Practice

Semantic communication (SemCom) promises to reduce transmitted payloads by conveying task-relevant meaning instead of raw bits. However, practical SemCom also incurs semantic metadata, control signaling, feedback, model or knowledge-base synchronization, and neural computation costs, which may offset semantic compression gains. This paper develops an overhead-aware analytical framework for quantifying the spectral-resource and energy costs of SemCom under equal task utility. The framework covers point-to-point transmission, user equipment (UE)-to-next-generation NodeB (gNB) uplink, and UE-to-UE communication under a single gNB, and derives closed-form break-even conditions with respect to payload size, semantic compression factor, model reuse, protocol overhead, and computation energy. Simulation results show that SemCom becomes spectrally beneficial only for sufficiently large payloads, while energy gains require larger payloads due to processing and synchronization overheads. The results also show that multi-user downlink is particularly favorable, as shared semantic overheads can be amortized across multiple UEs. These findings provide design guidance for realistic SemCom evaluation and standardization-oriented deployment.

cs.NI↗

arXiv2026

G2DP: Diffusion Planning with Spatio-Temporal Grid Guidance

In autonomous driving, diffusion-based planners have emerged as a promising paradigm for robust motion planning in dense and interactive traffic, as they can effectively model diverse driving behaviors. However, their inherent stochasticity often requires explicit guidance during denoising to ensure safety and route adherence for robust closed-loop execution. Existing guidance typically relies on sparse, entity-centric geometric queries or post-hoc refinement, yielding limited situational awareness and fragile performance in interactive scenes. To address this issue, we propose G2DP (Grid-Guided Diffusion Planning), a diffusion-based planner that directly enforces dense environmental constraints through inference-time guidance. Specifically, G2DP constructs a differentiable spatio-temporal cost volume by fusing probabilistic future occupancy distributions with a route-progress map. By formulating this volume as a continuous safety energy functional, it injects dense gradients directly into the denoising loop, actively steering trajectory generation toward collision-free and progress-optimal regions. Extensive closed-loop evaluations show that G2DP achieves state-of-the-art performance on nuPlan, outperforming the strongest imitation-learning baseline by +7.2 points in reactive score. It further maintains top scores in zero-shot transfers to interPlan and DeepScenario benchmarks, with collision avoidance improving by +10.15 over the unguided approach on interPlan. These results demonstrate that spatio-temporal cost grids serve as an effective representation for robust guidance in diffusion-based planning. Code is available at https://github.com/HangYuu/G2DP.

cs.RO↗

arXiv2026

SpaceDrive: Infusing Spatial Awareness into VLM-based Autonomous Driving

End-to-end autonomous driving methods built on vision language models (VLMs) have undergone rapid development driven by their universal visual understanding and strong reasoning capabilities obtained from the large-scale pretraining. However, we find that current VLMs struggle to understand fine-grained 3D spatial relationships which is a fundamental requirement for systems interacting with the physical world. To address this issue, we propose SpaceDrive, a spatial-aware VLM-based driving framework that treats spatial information as explicit positional encodings (PEs) instead of textual digit tokens, enabling joint reasoning over semantic and spatial representations. SpaceDrive employs a universal positional encoder to all 3D coordinates derived from multi-view depth estimation, historical ego-states, and text prompts. These 3D PEs are first superimposed to augment the corresponding 2D visual tokens. Meanwhile, they serve as a task-agnostic coordinate representation, replacing the digit-wise numerical tokens as both inputs and outputs for the VLM. This mechanism enables the model to better index specific visual semantics in spatial reasoning and directly regress trajectory coordinates rather than generating digit-by-digit, thereby enhancing planning accuracy. Extensive experiments validate that SpaceDrive achieves state-of-the-art open-loop performance on the nuScenes dataset and the second-best Driving Score of 78.02 on the Bench2Drive closed-loop benchmark over existing VLM-based methods. Code is available at: https://github.com/zhenghao2519/SpaceDrive.

cs.CV↗

arXiv2026

Shift & Drift: A Zero-Shot Benchmark for Generalizable and Robust Autonomous Driving Motion Planning

While closed-loop motion planners trained on large-scale, object-level datasets, e.g., nuPlan, demonstrate strong in-distribution (ID) performance, their generalization to novel urban topologies and recovery mechanisms following execution perturbations remain under-explored. To address this, we present Shift & Drift, a novel dual-track benchmark designed to rigorously stress-test motion planners across two critical axes of distribution shift: (1) The Semantic Shift Track leverages a novel conversion pipeline that transforms the aerial, DeepScenario Open 3D dataset into the nuPlan simulation framework. This enables zero-shot evaluation of planners trained on North American and Singaporean data against 1,182 scenarios spanning four German cities and the US city of San Francisco featuring dense pedestrian-cyclist interactions. (2) The State-Distribution Drift Track injects stochastic perturbations into the ego vehicle's dynamics to quantify robustness against compounding execution errors. Based on this, we systematically evaluate the failure modes of diverse planning paradigms under semantic and state-distribution shifts. While imitation learning methods achieve high scores in ID benchmarks, they exhibit significant failures under semantic shift, particularly in pedestrian-dense environments, and suffer from persistent drift when subjected to temporally correlated actuation noise. In contrast, the evaluated reinforcement-learning-based planner demonstrates more graceful degradation, maintaining higher safety and progress metrics across both tracks. Our findings reveal an empirical trade-off between imitation fidelity and closed-loop resilience, providing the community with a rigorous benchmark to evaluate progress toward reliable deployment.

cs.RO↗

arXiv2026

Glance-Say: Multimodal Human-Robot Collaboration and Intent Recognition via Sticky Glance

Gaze and speech are promising interaction modalities for individuals with motor impairments, yet robust intent recognition in multi-object environments remains challenging due to micro-saccades, semantic ambiguity, and viewpoint changes. This paper presents a multimodal interaction framework for assistive robotic manipulation. We propose a sticky-glance algorithm that stabilizes gaze-based intent by jointly accumulating geometric distance and directional evidence, enabling robust real-time target selection and switching. We further introduce Glance-Say, a gaze-speech interaction paradigm in which gaze specifies objects and speech specifies actions, together with a continuous shared-control scheme that provides high-readiness robot motion and human-in-the-loop feedback. Experiments demonstrate a tracking rate of 0.92 for moving targets, selection accuracy of 0.97 for static targets, and reduced task duration. These results indicate improved robustness, efficiency, and usability over representative interaction paradigms.

cs.RO↗

arXiv2026

Seizure-Semiology-Suite (S3): A Clinically Multimodal Dataset, Benchmark, and Models for Seizure Semiology Understanding

While Multimodal Large Language Models (MLLMs) have demonstrated remarkable proficiency in general video understanding, their capacity to interpret involuntary, and spatio-temporally evolving pathologic motor behaviors such as seizure semiology remains largely untested. To address this gap, we introduce Seizure-Semiology-Suite, a clinically grounded dataset and benchmark for fine-grained, structured seizure semiology understanding. The dataset includes 438 seizure videos annotated with over 35,000 dense labels covering 20 ILAE-defined semiological features. Building on this dataset, we propose a seven-task hierarchical benchmark that systematically evaluates MLLMs from low-level visual perception to temporal sequencing, narrative report generation, and seizure diagnosis. To enable clinically meaningful evaluation of generated reports, we further introduce the Report Quality Index for Seizure Semiology (Seizure-RQI). Extensive baselines across 11 open-weight MLLMs reveal systematic weaknesses in laterality reasoning, temporal localization, symptom sequencing, and clinically faithful reporting. We show that seizure-specific fine-tuning substantially improves performance across tasks, and that a two-stage neuro-symbolic framework achieves an F1 score of 0.96 on epileptic versus non-epileptic seizure classification. Seizure-Semiology-Suite establishes a rigorous benchmark for evaluating multimodal models in safety-critical medical video understanding and guides the development of clinically reliable, domain-adaptive multimodal intelligence.

cs.CV↗

arXiv2026

FAM-HRI: Foundation-Model Assisted Multi-Modal Human-Robot Interaction Combining Gaze and Speech

ffective Human-Robot Interaction (HRI) is crucial for enhancing accessibility and usability in real-world robotics applications. However, existing solutions often rely on gesture- only or language-only commands, making interaction inefficient and ambiguous, particularly for users with physical impairments. In this paper, we introduce FAM-HRI, an efficient multimodal framework for HRI that integrates language and gaze inputs via foundation models. By leveraging lightweight Meta ARIA glasses, our system captures real-time multimodal signals and utilizes large language models (LLMs) to fuse user intention with scene context, enabling intuitive and precise robot manipulation. Our method accurately determines the gaze fixation time interval, reducing noise caused by the gaze dynamic nature. Experimental evaluations demonstrate that FAM-HRI achieves a high success rate in task execution while maintaining a low interaction time, providing a practical solution for individuals with limited physical mobility or motor impairments. To support the community, we have released our system design, algorithms, and solutions at https://github.com/laiyuzhi/FAM-HRI.

cs.HC↗

arXiv2026

Can Multimodal Large Language Models Understand Pathologic Movements? A Pilot Study on Seizure Semiology

Multimodal Large Language Models (MLLMs) have demonstrated robust capabilities in recognizing everyday human activities, yet their potential for analyzing clinically significant involuntary movements in neurological disorders remains largely unexplored. This pilot study evaluates the capability of MLLMs for automated recognition of pathological movements in seizure videos. We assessed the zero-shot performance of state-of-the-art MLLMs on 20 ILAE-defined semiological features across 90 clinical seizure recordings. MLLMs outperformed fine-tuned Convolutional Neural Network (CNN) and Vision Transformer (ViT) baseline models on 13 of 18 features without task-specific training, demonstrating particular strength in recognizing salient postural and contextual features while struggling with subtle, high-frequency movements. Feature-targeted signal enhancement (facial cropping, pose estimation, audio denoising) improved performance on 10 of 20 features. Expert evaluation showed that 94.3 percent of MLLM-generated explanations for correctly predicted cases achieved at least 60 percent faithfulness scores, aligning with epileptologist reasoning. These findings demonstrate the potential of adapting general-purpose MLLMs for specialized clinical video analysis through targeted preprocessing strategies, offering a path toward interpretable, efficient diagnostic assistance. Our code is publicly available at https://github.com/LinaZhangUCLA/PathMotionMLLM.

cs.CV↗

arXiv2026

Latency-aware Human-in-the-Loop Reinforcement Learning for Semantic Communications

Semantic communication promises task-aligned transmission but must reconcile semantic fidelity with stringent latency guarantees in immersive and safety-critical services. This paper introduces a time-constrained human-in-the-loop reinforcement learning (TC-HITL-RL) framework that embeds human feedback, semantic utility, and latency control within a semantic-aware Open radio access network (RAN) architecture. We formulate semantic adaptation driven by human feedback as a constrained Markov decision process (CMDP) whose state captures semantic quality, human preferences, queue slack, and channel dynamics, and solve it via a primal--dual proximal policy optimization algorithm with action shielding and latency-aware reward shaping. The resulting policy preserves PPO-level semantic rewards while tightening the variability of both air-interface and near-real-time RAN intelligent controller processing budgets. Simulations over point-to-multipoint links with heterogeneous deadlines show that TC-HITL-RL consistently meets per-user timing constraints, outperforms baseline schedulers in reward, and stabilizes resource consumption, providing a practical blueprint for latency-aware semantic adaptation.

eess.SP↗

arXiv2025

A Multi-Year Urban Streetlight Imagery Dataset for Visual Monitoring and Spatio-Temporal Drift Detection

We present a large-scale, longitudinal visual dataset of urban streetlights captured by 22 fixed-angle cameras deployed across Bristol, U.K., from 2021 to 2025. The dataset contains over 526,000 images, collected hourly under diverse lighting, weather, and seasonal conditions. Each image is accompanied by rich metadata, including timestamps, GPS coordinates, and device identifiers. This unique real-world dataset enables detailed investigation of visual drift, anomaly detection, and MLOps strategies in smart city deployments. To promtoe seconardary analysis, we additionally provide a self-supervised framework based on convolutional variational autoencoders (CNN-VAEs). Models are trained separately for each camera node and for day/night image sets. We define two per-sample drift metrics: relative centroid drift, capturing latent space deviation from a baseline quarter, and relative reconstruction error, measuring normalized image-domain degradation. This dataset provides a realistic, fine-grained benchmark for evaluating long-term model stability, drift-aware learning, and deployment-ready vision systems. The images and structured metadata are publicly released in JPEG and CSV formats, supporting reproducibility and downstream applications such as streetlight monitoring, weather inference, and urban scene understanding. The dataset can be found at https://doi.org/10.5281/zenodo.17781192 and https://doi.org/10.5281/zenodo.17859120.

cs.CV↗

arXiv2025

AGO: Adaptive Grounding for Open World 3D Occupancy Prediction

Open-world 3D semantic occupancy prediction aims to generate a voxelized 3D representation from sensor inputs while recognizing both known and unknown objects. Transferring open-vocabulary knowledge from vision-language models (VLMs) offers a promising direction but remains challenging. However, methods based on VLM-derived 2D pseudo-labels with traditional supervision are limited by a predefined label space and lack general prediction capabilities. Direct alignment with pretrained image embeddings, on the other hand, often fails to achieve reliable performance because of inconsistent image and text representations in VLMs. To address these challenges, we propose AGO, a novel 3D occupancy prediction framework with adaptive grounding to handle diverse open-world scenarios. AGO first encodes surrounding images and class prompts into 3D and text embeddings, respectively, leveraging similarity-based grounding training with 3D pseudo-labels. Additionally, a modality adapter maps 3D embeddings into a space aligned with VLM-derived image embeddings, reducing modality gaps. Experiments on Occ3D-nuScenes show that AGO improves unknown object prediction in zero-shot and few-shot transfer while achieving state-of-the-art closed-world self-supervised performance, surpassing prior methods by 4.09 mIoU. Code is available at: https://github.com/EdwardLeeLPZ/AGO.

cs.CV↗

arXiv2025

Satellite Direct-to-Device from Low Earth Orbit: Techno-Economic Analysis of a Global Non-Terrestrial Network

Low Earth orbit (LEO) satellites and satellite direct-to-device (D2D) technology are at the heart of the next-generation global connectivity which promises direct access to space-based broadband services for unmodified 3GPP-compliant handsets. With a rapidly evolving ecosystem, it is important to evaluate the feasibility, cost-effectiveness, and profitability of these services. By assessing the technological aspects as well as economic implications, stakeholders can make informed decisions about investment, development, and deployment strategies. This paper presents a comprehensive techno-economic analysis (TEA) framework for evaluating LEO-based satellite D2D systems. The framework integrates a global satellite constellation model, radio propagation aspects including atmospheric and rainfall attenuation models compliant with ITU-R recommendations, 3GPP-compliant capacity calculations, realistic global population data, and an all-encompassing cost model accounting for both capital and operational expenses associated with space and ground segments. Further, the framework evaluates three different architectural options for realizing a global non-terrestrial network (NTN) for satellite D2D services. With an emphasis on reproducibility, the framework has been implemented through significant enhancements to an open-source tool. The economic assessment reveals that global satellite D2D services can be provided at a monthly cost per subscriber which is comparable to terrestrial services while achieving a positive return on investment (ROI). Moreover, the results show the potential of Open RAN technology for realizing cost-effective satellite D2D services.

cs.NI↗

arXiv2025

Building the Self-Improvement Loop: Error Detection and Correction in Goal-Oriented Semantic Communications

Error detection and correction are essential for ensuring robust and reliable operation in modern communication systems, particularly in complex transmission environments. However, discussions on these topics have largely been overlooked in semantic communication (SemCom), which focuses on transmitting meaning rather than symbols, leading to significant improvements in communication efficiency. Despite these advantages, semantic errors -- stemming from discrepancies between transmitted and received meanings -- present a major challenge to system reliability. This paper addresses this gap by proposing a comprehensive framework for detecting and correcting semantic errors in SemCom systems. We formally define semantic error, detection, and correction mechanisms, and identify key sources of semantic errors. To address these challenges, we develop a Gaussian process (GP)-based method for latent space monitoring to detect errors, alongside a human-in-the-loop reinforcement learning (HITL-RL) approach to optimize semantic model configurations using user feedback. Experimental results validate the effectiveness of the proposed methods in mitigating semantic errors under various conditions, including adversarial attacks, input feature changes, physical channel variations, and user preference shifts. This work lays the foundation for more reliable and adaptive SemCom systems with robust semantic error management techniques.

cs.NI↗

arXiv2025

Anomaly Detection in Offshore Open Radio Access Network Using Long Short-Term Memory Models on a Novel Artificial Intelligence-Driven Cloud-Native Data Platform

The Radio Access Network (RAN) is a critical component of modern telecommunications infrastructure, currently evolving towards disaggregated and open architectures. These advancements are pivotal for integrating intelligent, data-driven applications aimed at enhancing network reliability and operational autonomy through the introduction of cognitive capabilities, as exemplified by the emerging Open Radio Access Network (O-RAN) standards. Despite its potential, the nascent nature of O-RAN technology presents challenges, primarily due to the absence of mature operational standards. This complicates the management of data and intelligent applications, particularly when integrating with traditional network management and operational support systems. Divergent vendor-specific design approaches further hinder migration and limit solution reusability. These challenges are compounded by a skills gap in telecommunications business-oriented engineering, which remains a key barrier to effective O-RAN deployment and intelligent application development. To address these challenges, Boldyn Networks developed a novel cloud-native data analytics platform, specifically designed to support scalable AI integration within O-RAN deployments. This platform underwent rigorous testing in real-world scenarios, and applied advanced Artificial Intelligence (AI) techniques to improve operational efficiency and customer experience. Implementation involved adopting Development Operations (DevOps) practices, leveraging data lakehouse architectures tailored for AI applications, and employing sophisticated data engineering strategies. The platform successfully addresses connectivity challenges inherent in real-world offshore windfarm deployments using Long Short-Term Memory (LSTM) models for anomaly detection in network connectivity.

cs.NI↗