arXiv subjects

Yiqin Wang

Publications and source records attributed to Yiqin Wang.

At least 19 recordsLinked to original sources

arXiv2026

FoodMonitor: Benchmarking MLLMs for Explainable Compliance Analysis

As AI-powered compliance monitoring becomes increasingly important in public governance and industrial safety, the ability to provide verifiable evidence and traceable accountability signals is essential. However, existing video anomaly detection datasets focus on event-level binary classification, lacking the rule-driven, explainable analysis required for real-world compliance scenarios. We introduce FoodMonitor, a benchmark for explainable compliance analysis in commercial kitchen surveillance. FoodMonitor comprises 477 video clips with 3,307 violation annotations across a dual-channel design covering both person-level and environment-level violations. Each annotation specifies which rule was violated, what non-compliant behavior occurred, and who committed it with frame-level bounding boxes. We establish a unified evaluation protocol with a two-stage matching mechanism that separately assesses spatial localization and semantic understanding, along with a composite metric ($C_{\text{score}}$) that balances environment and person detection performance. Systematic evaluation of several state-of-the-art multimodal large language models reveals that the best-performing model achieves only 0.360 $C_{\text{score}}$, with spatial localization and fine-grained rule understanding emerging as the primary bottlenecks. Our analysis identifies two distinct failure modes: localization-dominated errors and semantics-dominated errors, providing diagnostic insights for future model development.

cs.CV↗

arXiv2026

Robust Spatiotemporal Motion Planning for Multi-Agent Autonomous Racing via Topological Gap Identification and Accelerated MPC

High-speed multi-agent autonomous racing demands robust spatiotemporal planning and precise control under strict computational limits. Current methods often oversimplify interactions or abandon strict kinematic constraints. We resolve this by proposing a Topological Gap Identification and Accelerated MPC framework. By predicting opponent behaviors via SGPs, our method constructs dynamic occupancy corridors to robustly select optimal overtaking gaps. We ensure strict kinematic feasibility using a Linear Time-Varying MPC powered by a customized Pseudo-Transient Continuation (PTC) solver for high-frequency execution. Experimental results on the F1TENTH platform show that our method significantly outperforms state-of-the-art baselines: it reduces total maneuver time by 51.6% in sequential scenarios, consistently maintains an overtaking success rate exceeding 81% in dense bottlenecks, and lowers average computational latency by 20.3%, pushing the boundaries of safe and high-speed autonomous racing.

cs.RO↗

arXiv2026

Vision-Augmented On-Track System Identification for Autonomous Racing via Attention-Based Priors and Iterative Neural Correction

Operating autonomous vehicles at the absolute limits of handling requires precise, real-time identification of highly non-linear tire dynamics. However, traditional online optimization methods suffer from "cold-start" initialization failures and struggle to model high-frequency transient dynamics. To address these bottlenecks, this paper proposes a novel vision-augmented, iterative system identification framework. First, a lightweight CNN (MobileNetV3) translates visual road textures into a continuous heuristic friction prior, providing a robust "warm-start" for parameter optimization. Next, a S4 model captures complex temporal dynamic residuals, circumventing the memory and latency limitations of traditional MLPs and RNNs. Finally, a derivative-free Nelder-Mead algorithm iteratively extracts physically interpretable Pacejka tire parameters via a hybrid virtual simulation. Co-simulation in CarSim demonstrates that the lightweight vision backbone reduces friction estimation error by 76.1 using 85 fewer FLOPs, accelerating cold-start convergence by 71.4. Furthermore, the S4-augmented framework improves parameter extraction accuracy and decreases lateral force RMSE by over 60 by effectively capturing complex vehicle dynamics, demonstrating superior performance compared to conventional neural architectures.

cs.RO↗

arXiv2026

BrainRVQ: A High-Fidelity EEG Foundation Model via Dual-Domain Residual Quantization and Hierarchical Autoregression

Developing foundation models for electroencephalography (EEG) remains challenging due to the signal's low signal-to-noise ratio and complex spectro-temporal non-stationarity. Existing approaches often overlook the hierarchical latent structure inherent in neural dynamics, leading to suboptimal reconstruction of fine-grained information. In this work, we propose BrainRVQ, a general-purpose EEG foundation model pre-trained on a large-scale corpus of clinical EEG data. Unlike standard masked modeling, BrainRVQ features a Dual-Domain Residual Vector Quantization (DD-RVQ) tokenizer that disentangles temporal waveforms and spectral patterns into hierarchical discrete codes. We further introduce a hierarchical autoregressive pre-training objective that learns to reconstruct these codes in a coarse-to-fine manner, utilizing an importance-guided curriculum masking strategy to prioritize information-rich neural events over background noise. Extensive experiments across 8 diverse downstream datasets demonstrate that BrainRVQ consistently outperforms state-of-the-art baselines, validating its effectiveness in learning robust and generalizable neural representations. Our code and model weights are available:https://github.com/keqicmz/BrainRVQ

eess.SP↗

arXiv2025

MCTR: Midpoint Corrected Triangulation for Autonomous Racing via Digital Twin Simulation in CARLA

In autonomous racing, reactive controllers eliminate the computational burden of the full See-Think-Act autonomy stack by directly mapping sensor inputs to control actions. This bypasses the need for explicit localization and trajectory planning. A widely adopted baseline in this category is the Follow-The-Gap method, which performs trajectory planning using LiDAR data. Building on FTG, the Delaunay Triangulation-based Racing algorithm introduces further enhancements. However, DTR's use of circumcircles for trajectory generation often results in insufficiently smooth paths, ultimately degrading performance. Additionally, the commonly used F1TENTH-simulator for autonomous racing competitions lacks support for 3D LiDAR perception, limiting its effectiveness in realistic testing. To address these challenges, this work proposes the MCTR algorithm. MCTR improves trajectory smoothness through the use of Curvature Corrected Moving Average and implements a digital twin system within the CARLA simulator to validate the algorithm's robustness under 3D LiDAR perception. The proposed algorithm has been thoroughly validated through both simulation and real-world vehicle experiments.

cs.RO↗

arXiv2025

Flash-VStream: Efficient Real-Time Understanding for Long Video Streams

Benefiting from the advances in large language models and cross-modal alignment, existing multimodal large language models have achieved prominent performance in image and short video understanding. However, the understanding of long videos is still challenging, as their long-context nature results in significant computational and memory overhead. Most existing work treats long videos in the same way as short videos, which is inefficient for real-world applications and hard to generalize to even longer videos. To address these issues, we propose Flash-VStream, an efficient video language model capable of processing extremely long videos and responding to user queries in real time. Particularly, we design a Flash Memory module, containing a low-capacity context memory to aggregate long-context temporal information and model the distribution of information density, and a high-capacity augmentation memory to retrieve detailed spatial information based on this distribution. Compared to existing models, Flash-VStream achieves significant reductions in inference latency. Extensive experiments on long video benchmarks and comprehensive video benchmarks, i.e., EgoSchema, MLVU, LVBench, MVBench and Video-MME, demonstrate the state-of-the-art performance and outstanding efficiency of our method. Code is available at https://github.com/IVGSZ/Flash-VStream.

cs.CV↗

arXiv2025

Kimi-VL Technical Report

We present Kimi-VL, an efficient open-source Mixture-of-Experts (MoE) vision-language model (VLM) that offers advanced multimodal reasoning, long-context understanding, and strong agent capabilities - all while activating only 2.8B parameters in its language decoder (Kimi-VL-A3B). Kimi-VL demonstrates strong performance across challenging domains: as a general-purpose VLM, Kimi-VL excels in multi-turn agent tasks (e.g., OSWorld), matching flagship models. Furthermore, it exhibits remarkable capabilities across diverse challenging vision language tasks, including college-level image and video comprehension, OCR, mathematical reasoning, and multi-image understanding. In comparative evaluations, it effectively competes with cutting-edge efficient VLMs such as GPT-4o-mini, Qwen2.5-VL-7B, and Gemma-3-12B-IT, while surpassing GPT-4o in several key domains. Kimi-VL also advances in processing long contexts and perceiving clearly. With a 128K extended context window, Kimi-VL can process diverse long inputs, achieving impressive scores of 64.5 on LongVideoBench and 35.1 on MMLongBench-Doc. Its native-resolution vision encoder, MoonViT, further allows it to see and understand ultra-high-resolution visual inputs, achieving 83.2 on InfoVQA and 34.5 on ScreenSpot-Pro, while maintaining lower computational cost for common tasks. Building upon Kimi-VL, we introduce an advanced long-thinking variant: Kimi-VL-Thinking-2506. Developed through long chain-of-thought (CoT) supervised fine-tuning (SFT) and reinforcement learning (RL), the latest model exhibits strong long-horizon reasoning capabilities (64.0 on MMMU, 46.3 on MMMU-Pro, 56.9 on MathVision, 80.1 on MathVista, 65.2 on VideoMMMU) while obtaining robust general abilities. Code and models are publicly accessible at https://github.com/MoonshotAI/Kimi-VL.

cs.CV↗

arXiv2025

A Geography-Inspired and Self-Adaptive Clustering Algorithm: A Study in Channel Measurement

The phenomenon that multi-path components (MPCs) arrive in clusters has been verified by channel measurements, and is widely adopted by cluster-based channel models. As a crucial intermediate processing step, MPC clustering bridges raw data in channel measurement and cluster characteristics for channel modeling. In this paper, a physical-interpretable and self-adaptive MPC clustering algorithm is proposed, which can locate both single-point and wide-spread scatterers without prior knowledge. Inspired by the concept in geography, a novel metaphor that interprets features of MPC attributes in the power-delay-angle profile (PDAP) as topographic concepts is developed. In light of the interpretation, the proposed algorithm disassembles the PDAP by constructing contour lines and identifying characteristic points that indicate the skeleton of MPC clusters, which are fitted by analytical models that associate MPCs with physical scatterer locations. Besides, a new clustering performance index, the power gradient consistency index, is proposed. Calculated as the weighted Spearman correlation coefficient between the power and the distance to the center, the index captures the intrinsic property of MPC clusters that the dominant high-power path is surrounded by lower-power paths. The performance of the proposed algorithm is analyzed and compared with the counterparts of conventional clustering algorithms based on the channel measurement conducted in an outdoor scenario. The proposed algorithm performs better in average Silhouette index and weighted Spearman correlation coefficient, and the average root mean square error (RMSE) of the estimated scatterer location is 0.1 m.

cs.IT↗

arXiv2024

Hierarchical Memory for Long Video QA

This paper describes our champion solution to the LOVEU Challenge @ CVPR'24, Track 1 (Long Video VQA). Processing long sequences of visual tokens is computationally expensive and memory-intensive, making long video question-answering a challenging task. The key is to compress visual tokens effectively, reducing memory footprint and decoding latency, while preserving the essential information for accurate question-answering. We adopt a hierarchical memory mechanism named STAR Memory, proposed in Flash-VStream, that is capable of processing long videos with limited GPU memory (VRAM). We further utilize the video and audio data of MovieChat-1K training set to fine-tune the pretrained weight released by Flash-VStream, achieving 1st place in the challenge. Code is available at project homepage https://invinciblewyq.github.io/vstream-page .

cs.CV↗

arXiv2024

Ponder & Press: Advancing Visual GUI Agent towards General Computer Control

Most existing GUI agents typically depend on non-vision inputs like HTML source code or accessibility trees, limiting their flexibility across diverse software environments and platforms. Current multimodal large language models (MLLMs), which excel at using vision to ground real-world objects, offer a potential alternative. However, they often struggle with accurately localizing GUI elements -- a critical requirement for effective GUI automation -- due to the semantic gap between real-world objects and GUI elements. In this work, we introduce Ponder & Press, a divide-and-conquer framework for general computer control using only visual input. Our approach combines an general-purpose MLLM as an 'interpreter', responsible for translating high-level user instructions into detailed action descriptions, with a GUI-specific MLLM as a 'locator' that precisely locates GUI elements for action placement. By leveraging a purely visual input, our agent offers a versatile, human-like interaction paradigm applicable to a wide range of applications. Ponder & Press locator outperforms existing models by +22.5% on the ScreenSpot GUI grounding benchmark. Both offline and interactive agent benchmarks across various GUI environments -- including web pages, desktop software, and mobile UIs -- demonstrate that Ponder & Press framework achieves state-of-the-art performance, highlighting the potential of visual GUI agents. Refer to the project homepage https://invinciblewyq.github.io/ponder-press-page/

cs.CV↗

arXiv2024

Coarse Correspondences Boost Spatial-Temporal Reasoning in Multimodal Language Model

Multimodal language models (MLLMs) are increasingly being applied in real-world environments, necessitating their ability to interpret 3D spaces and comprehend temporal dynamics. Current methods often rely on specialized architectural designs or task-specific fine-tuning to achieve this. We introduce Coarse Correspondences, a simple lightweight method that enhances MLLMs' spatial-temporal reasoning with 2D images as input, without modifying the architecture or requiring task-specific fine-tuning. Our method uses a lightweight tracking model to identify primary object correspondences between frames in a video or across different image viewpoints, and then conveys this information to MLLMs through visual prompting. We demonstrate that this simple training-free approach brings substantial gains to GPT4-V/O consistently on four benchmarks that require spatial-temporal reasoning, including +20.5\% improvement on ScanQA, +9.7\% on OpenEQA's episodic memory subset, +6.0\% on the long-form video benchmark EgoSchema, and +11\% on the R2R navigation benchmark. Additionally, we show that Coarse Correspondences can also enhance open-source MLLMs' spatial reasoning (by +6.9\% on ScanQA) when applied in both training and inference and that the improvement can generalize to unseen datasets such as SQA3D (+3.1\%). Taken together, we show that Coarse Correspondences effectively and efficiently boosts models' performance on downstream tasks requiring spatial-temporal reasoning.

cs.CV↗

arXiv2024

Movable Antennas: Channel Measurement, Modeling, and Performance Evaluation

Since decades ago, multi-antenna has become a key enabling technology in the evolution of wireless communication systems. In contrast to conventional multi-antenna systems that contain antennas at fixed positions, position-flexible antenna systems have been proposed to fully utilize the spatial variation of wireless channels. In this paper, movable antenna (MA) systems are analyzed from channel measurement, modeling, position optimization to performance evaluation. First, a broadband channel measurement system with physical MAs is developed, for which the extremely high movable resolution reaches 0.02 mm. A practical two-ray model is constructed based on the channel measurement for a two-dimensional movable antenna system across 32$\times$32 planar port positions at 300 GHz. In light of the measurement results, spatial-correlated channel models for the two-dimensional MA system are proposed, which are statistically parameterized by the covariance matrix of measured channels. Finally, the signal-to-interference-and-noise ratio (SINR)-maximized position selection algorithm is proposed, which achieves 99% of the optimal performance. The performance of different MA systems in terms of spectral efficiency are evaluated and compared for both planar and linear MA systems. Extensive results demonstrate the advantage of MAs over fixed-position antennas in coping with the multi-path fading and improving the spectral efficiency by 10% in a 300 GHz measured channel.

cs.IT↗

arXiv2024

220 GHz Urban Microcell Channel Measurement and Characterization on a University Campus

Owning abundant bandwidth resources, the Terahertz (THz) band (0.1-10~THz) is envisioned as a key technology to realize ultra-high-speed communications in 6G and beyond wireless networks. To realize reliable THz communications in urban microcell (UMi) environments, propagation analysis and channel characterization are still insufficient. In this paper, channel measurement campaigns are conducted in a UMi scenario at 220~GHz, using a correlation-based time domain channel sounder. 24 positions are measured along a road on the university campus, with distances ranging from 34~m to 410~m. Based on the measurement results, the spatial consistency and interaction of THz waves to the surrounding environments are analyzed. Moreover, the additional loss due to foliage blockage is calculated and an average value of 16.7~dB is observed. Furthermore, a full portrait of channel characteristics, including path loss, shadow fading, K-factor, delay and angular spreads, as well as cluster parameters, is calculated and analyzed. Specifically, an average K-factor value of 17.5 dB is measured in the line-of-sight (LoS) case, which is nearly two times larger than the extrapolated values from the 3GPP standard, revealing weak multipath effects in the THz band. Additionally, 2.5 clusters on average are observed in the LoS case, around one fifth of what is defined in the 3GPP model, which uncovers the strong sparsity in THz UMi. The results and analysis in this work can offer guidance for system design for future THz UMi networks.

cs.IT↗

arXiv2024

Flash-VStream: Memory-Based Real-Time Understanding for Long Video Streams

Benefiting from the advancements in large language models and cross-modal alignment, existing multi-modal video understanding methods have achieved prominent performance in offline scenario. However, online video streams, as one of the most common media forms in the real world, have seldom received attention. Compared to offline videos, the 'dynamic' nature of online video streams poses challenges for the direct application of existing models and introduces new problems, such as the storage of extremely long-term information, interaction between continuous visual content and 'asynchronous' user questions. Therefore, in this paper we present Flash-VStream, a video-language model that simulates the memory mechanism of human. Our model is able to process extremely long video streams in real-time and respond to user queries simultaneously. Compared to existing models, Flash-VStream achieves significant reductions in inference latency and VRAM consumption, which is intimately related to performing understanding of online streaming video. In addition, given that existing video understanding benchmarks predominantly concentrate on offline scenario, we propose VStream-QA, a novel question answering benchmark specifically designed for online video streaming understanding. Comparisons with popular existing methods on the proposed benchmark demonstrate the superiority of our method for such challenging setting. To verify the generalizability of our approach, we further evaluate it on existing video understanding benchmarks and achieves state-of-the-art performance in offline scenarios as well. All code, models, and datasets are available at the https://invinciblewyq.github.io/vstream-page/

cs.CV↗

arXiv2024

Cross Far- and Near-Field Channel Measurement and Modeling in Extremely Large-scale Antenna Array (ELAA) Systems

Technologies like ultra-massive multiple-input-multiple-output (UM-MIMO) and reconfigurable intelligent surfaces (RISs) are of special interest to meet the key performance indicators of future wireless systems including ubiquitous connectivity and lightning-fast data rates. One of their common features, the extremely large-scale antenna array (ELAA) systems with hundreds or thousands of antennas, give rise to near-field (NF) propagation and bring new challenges to channel modeling and characterization. In this paper, a cross-field channel model for ELAA systems is proposed, which improves the statistical model in 3GPP TR 38.901 by refining the propagation path with its first and last bounces and differentiating the characterization of parameters like path loss, delay, and angles in near- and far-fields. A comprehensive analysis of cross-field boundaries and closed-form expressions of corresponding NF or FF parameters are provided. Furthermore, cross-field experiments carried out in a typical indoor scenario at 300 GHz verify the variation of MPC parameters across the antenna array, and demonstrate the distinction of channels between different antenna elements. Finally, detailed generation procedures of the cross-field channel model are provided, based on which simulations and analysis on NF probabilities and channel coefficients are conducted for $4\times4$, $8\times8$, $16\times16$, and $9\times21$ uniform planar arrays at different frequency bands.

cs.IT↗

arXiv2024

Correlation-based Dual-band THz Channel Measurements and Characterization in a Laboratory

The Terahertz band, spanning from 0.1~THz to 10~THz, is envisioned as a key technology to realize ultra-high data rates in the 6G and beyond mobile communication systems, due to its abundant bandwidth resource. However, to realize THz communications, one substantial step is to fully understand the THz channels, which relies on extensive channel measurements. In this paper, using a correlation-based time domain channel sounder, measurement campaigns are conducted in a laboratory at 140~GHz and 220~GHz. In the data post-processing procedures, the time drift of clock signals is corrected using a linear interpolation/extrapolation method. Based on the measured results, the main objects that provide significant once-scattering clusters are found, based on which the scattering losses are calculated and analyzed. Furthermore, the channel characteristics, including path loss, shadow fading, K-factor, etc. are calculated and compared to 3GPP standard values. The propagation analysis and channel characteristics are helpful to study channel modeling and guide system design for THz communications.

cs.IT↗

arXiv2024

Far- and Near-Field Channel Measurements and Characterization in the Terahertz Band Using a Virtual Antenna Array

Extremely large-scale antenna array (ELAA) technologies consisting of ultra-massive multiple-input-multiple-output (UM-MIMO) or reconfigurable intelligent surfaces (RISs), are emerging to meet the demand of wireless systems in sixth-generation and beyond communications for enhanced coverage and extreme data rates up to Terabits per second. For ELAA operating at Terahertz (THz) frequencies, the Rayleigh distance expands, and users are likely to be located in both far-field (FF) and near-field (NF) regions. On one hand, new features like NF propagation and spatial non-stationarity need to be characterized. On the other hand, the transition of properties near the FF and NF boundary is worth exploring. In this paper, a complete experimental analysis of far- and near-field channel characteristics using a THz virtual antenna array is provided based on measurement of the multi-input-single-output channel with the virtual uniform planar array (UPA) structure of at most 4096 elements. In particular, non-linear phase change is observed in the NF, and the Rayleigh criterion regarding the maximum phase error is verified. Then, a new cross-field path loss model is proposed, which characterizes the power change at antenna elements in the UPA and is compatible with both FF and NF cases.

cs.IT↗

arXiv2023

Channel Measurement and Coverage Analysis for NIRS-Aided THz Communications in Indoor Environments

Due to large reflection and diffraction losses in the THz band, it is arguable to achieve reliable links in the none-line-of-sight (NLoS) cases. Intelligent reflecting surfaces, although are expected to solve the blockage problem and enhance the system connectivity, suffer from fabrication difficulty and operation complexity. In this work, non-intelligent reflecting surfaces (NIRS), which are simply made of costless metal foils and have no signal configuration capability, are adopted to enhance the signal strength and coverage in the THz band. Channel measurements are conducted in typical indoor scenarios at 306-321 GHz and 356-371 GHz bands to validate the effectiveness of the NIRS. Results measured with NIRS in different sizes show that large NIRS performs much better than small NIRS. Furthermore, by invoking the NIRS, the additional reflection loss can be reduced by more than 10~dB and the coverage ratio is increased by up to 39$\%$ for a 10~dB signal-to-noise ratio (SNR) threshold.

cs.IT↗