arXiv subjects

Hongyu Li

Publications and source records attributed to Hongyu Li.

At least 73 records · Page 4Linked to original sources

arXiv2025

Lossy Beyond Diagonal Reconfigurable Intelligent Surfaces: Modeling and Optimization

Beyond diagonal reconfigurable intelligent surface (BD-RIS) has emerged as an advancement and generalization of the conventional diagonal RIS (D-RIS) by introducing tunable interconnections between RIS elements, enabling smarter wave manipulation and enlarged coverage. While BD-RIS has demonstrated advantages over D-RIS in various aspects, most existing works rely on the assumption of a lossless model, leaving practical considerations unaddressed. This paper thus proposes a lossy BD-RIS model and develops corresponding optimization algorithms for various BD-RIS-aided communication systems. First, by leveraging admittance parameter analysis, we model each tunable admittance based on a lumped circuit with losses and derive an expression of a circle characterizing the real and imaginary parts of each tunable admittance. We then consider the received signal power maximization in single-user single-input single-output (SISO) systems with the proposed lossy BD-RIS model. To solve the optimization problem, we design an effective algorithm by carefully exploiting the problem structure. Specifically, an alternating direction method of multipliers (ADMM) framework is custom-designed to deal with the complicated constraints associated with lossy BD-RIS. Furthermore, we extend the proposed algorithmic framework to more general multiuser multiple-input single-output (MU-MISO) systems, where the transmit precoder and BD-RIS scattering matrix are jointly designed to maximize the sum-rate of the system. Finally, simulation results demonstrate that all BD-RIS architectures still outperform D-RIS in the presence of losses, but the optimal BD-RIS architectures in the lossless case are not necessarily optimal in the lossy case, e.g. group-connected BD-RIS can outperform fully- and tree-connected BD-RISs in SISO systems with relatively high losses, whereas the opposite always holds true in the lossless case.

eess.SP↗

arXiv2025

LongCat-Video Technical Report

Video generation is a critical pathway toward world models, with efficient long video inference as a key capability. Toward this end, we introduce LongCat-Video, a foundational video generation model with 13.6B parameters, delivering strong performance across multiple video generation tasks. It particularly excels in efficient and high-quality long video generation, representing our first step toward world models. Key features include: Unified architecture for multiple tasks: Built on the Diffusion Transformer (DiT) framework, LongCat-Video supports Text-to-Video, Image-to-Video, and Video-Continuation tasks with a single model; Long video generation: Pretraining on Video-Continuation tasks enables LongCat-Video to maintain high quality and temporal coherence in the generation of minutes-long videos; Efficient inference: LongCat-Video generates 720p, 30fps videos within minutes by employing a coarse-to-fine generation strategy along both the temporal and spatial axes. Block Sparse Attention further enhances efficiency, particularly at high resolutions; Strong performance with multi-reward RLHF: Multi-reward RLHF training enables LongCat-Video to achieve performance on par with the latest closed-source and leading open-source models. Code and model weights are publicly available to accelerate progress in the field.

cs.CV↗

arXiv2025

Global Optimal Closed-Form Solutions for Intelligent Surfaces With Mutual Coupling: Is Mutual Coupling Detrimental or Beneficial?

Reconfigurable Intelligent Surface (RIS) is a breakthrough technology enabling the dynamic control of the propagation environment in wireless communications through programmable surfaces. To improve the flexibility of conventional diagonal RIS (D-RIS), beyond diagonal RIS (BD-RIS) has emerged as a family of more general RIS architectures. However, D-RIS and BD-RIS have been commonly explored neglecting mutual coupling effects, while the global optimization of RIS with mutual coupling, its performance limits, and scaling laws remain unexplored. This study addresses these gaps by deriving global optimal closed-form solutions for BD-RIS with mutual coupling to maximize the channel gain, specifically fully- and tree-connected RISs. Besides, we provide the expression of the maximum channel gain achievable in the presence of mutual coupling and its scaling law in closed form. By using the derived scaling laws, we analytically prove that mutual coupling increases the channel gain on average under Rayleigh fading channels. Our theoretical analysis, confirmed by numerical simulations, shows that both fully- and tree-connected RISs with mutual coupling achieve the same channel gain upper bound when optimized with the proposed global optimal solutions. Furthermore, we observe that a mutual coupling-unaware optimization of RIS can cause a channel gain degradation of up to 5 dB.

cs.IT↗

arXiv2025

NovaFlow: Zero-Shot Manipulation via Actionable Flow from Generated Videos

Enabling robots to execute novel manipulation tasks zero-shot is a central goal in robotics. Most existing methods assume in-distribution tasks or rely on fine-tuning with embodiment-matched data, limiting transfer across platforms. We present NovaFlow, an autonomous manipulation framework that converts a task description into an actionable plan for a target robot without any demonstrations. Given a task description, NovaFlow synthesizes a video using a video generation model and distills it into 3D actionable object flow using off-the-shelf perception modules. From the object flow, it computes relative poses for rigid objects and realizes them as robot actions via grasp proposals and trajectory optimization. For deformable objects, this flow serves as a tracking objective for model-based planning with a particle-based dynamics model. By decoupling task understanding from low-level control, NovaFlow naturally transfers across embodiments. We validate on rigid, articulated, and deformable object manipulation tasks using a table-top Franka arm and a Spot quadrupedal mobile robot, and achieve effective zero-shot execution without demonstrations or embodiment-specific training. Project website: https://novaflow.lhy.xyz/.

cs.RO↗

arXiv2025

Mini-Omni-Reasoner: Token-Level Thinking-in-Speaking in Large Speech Models

Reasoning is essential for effective communication and decision-making. While recent advances in LLMs and MLLMs have shown that incorporating explicit reasoning significantly improves understanding and generalization, reasoning in LSMs remains in a nascent stage. Early efforts attempt to transfer the "Thinking-before-Speaking" paradigm from textual models to speech. However, this sequential formulation introduces notable latency, as spoken responses are delayed until reasoning is fully completed, impairing real-time interaction and communication efficiency. To address this, we propose Mini-Omni-Reasoner, a framework that enables reasoning within speech via a novel "Thinking-in-Speaking" formulation. Rather than completing reasoning before producing any verbal output, Mini-Omni-Reasoner interleaves silent reasoning tokens with spoken response tokens at the token level. This design allows continuous speech generation while embedding structured internal reasoning, leveraging the model's high-frequency token processing capability. Although interleaved, local semantic alignment is enforced to ensure that each response token is informed by its preceding reasoning. To support this framework, we introduce Spoken-Math-Problems-3M, a large-scale dataset tailored for interleaved reasoning and response. The dataset ensures that verbal tokens consistently follow relevant reasoning content, enabling accurate and efficient learning of speech-coupled reasoning. Built on a hierarchical Thinker-Talker architecture, Mini-Omni-Reasoner delivers fluent yet logically grounded spoken responses, maintaining both naturalness and precision. On the Spoken-MQA benchmark, it achieves a +19.1% gain in arithmetic reasoning and +6.4% in contextual understanding, with shorter outputs and zero decoding latency.

cs.CL↗

arXiv2025

Non-reciprocal Beyond Diagonal RIS: Sum-Rate Maximization in Full-Duplex Communications

Reconfigurable intelligent surface (RIS) has been envisioned as a key technology in future wireless communication networks to enable smart radio environment. To further enhance the passive beamforming capability of RIS, beyond diagonal (BD)-RIS has been proposed considering reconfigurable interconnections among different RIS elements. BD-RIS has a unique feature that cannot be enabled by conventional diagonal RIS; it can be realized by non-reciprocal circuits and thus enables an asymmetric scattering matrix. This feature provides the capability to break the wireless channel reciprocity, and has the potential to benefit full-duplex (FD) systems. In this paper, we model the BD RIS-assisted FD systems, where the impact of BD-RIS non-reciprocity and that of structural scattering, which refers to the specular reflection generated by RIS when the RIS is turned OFF, are explicitly captured. To assess the benefits of non-reciprocal BD-RIS, we optimise the scattering matrix, precoder and combiner to maximize the DL and UL sum-rates in the FD system. To tackle this optimization problem, we propose an iterative algorithm based on block coordination descent (BCD) and penalty dual decomposition (PDD). Numerical results demonstrate surprising benefits of non-reciprocal BD-RIS that it can achieve much higher DL and UL sum-rates in the FD scenario than reciprocal BD-RIS and conventional diagonal RIS.

eess.SP↗

arXiv2025

MIMO Channel Shaping and Rate Maximization Using Beyond-Diagonal RIS

This paper investigates the limits to which a passive Reconfigurable Intelligent Surface (RIS) can reshape a point-to-point Multiple-Input Multiple-Output (MIMO) channel in terms of singular values and their functions (e.g., achievable rate and harvestable power) for improved wireless performance. We depart from the Diagonal (D) scattering model and adopt a Beyond-Diagonal (BD) model that exploits element-wise connections for passive signal amplitude and phase manipulation. Specifically, analytical tight bounds are derived under typical RIS deployment scenarios to unveil the channel shaping potentials of BD-RIS regarding communication Degrees of Freedom (DoF), singular value spread, power gain, and capacity. An efficient numerical method is then proposed to optimize BD-RIS for any locally Lipschitz function of channel singular values, and showcased to characterize the achievable singular value region. As a side product, we tackle BD-RIS-aided MIMO rate maximization problem by a local-optimal Alternating Optimization (AO) approach and a low-complexity shaping approach. Results show that BD-RIS significantly improves the dynamic range of channel singular values and the tradeoff in manipulating them, thus offering enhanced data rate, harvestable power, and physical-layer security. These advantages become more pronounced when the number of RIS elements, group size, or MIMO dimensions increase. Of particular interest, BD-RIS is shown to activate multi-stream transmission and achieve the asymptotic DoF at much lower transmit power than D-RIS thanks to its proficiency in channel shaping.

eess.SP↗

arXiv2025

V-HOP: Visuo-Haptic 6D Object Pose Tracking

Humans naturally integrate vision and haptics for robust object perception during manipulation. The loss of either modality significantly degrades performance. Inspired by this multisensory integration, prior object pose estimation research has attempted to combine visual and haptic/tactile feedback. Although these works demonstrate improvements in controlled environments or synthetic datasets, they often underperform vision-only approaches in real-world settings due to poor generalization across diverse grippers, sensor layouts, or sim-to-real environments. Furthermore, they typically estimate the object pose for each frame independently, resulting in less coherent tracking over sequences in real-world deployments. To address these limitations, we introduce a novel unified haptic representation that effectively handles multiple gripper embodiments. Building on this representation, we introduce a new visuo-haptic transformer-based object pose tracker that seamlessly integrates visual and haptic input. We validate our framework in our dataset and the Feelsight dataset, demonstrating significant performance improvement on challenging sequences. Notably, our method achieves superior generalization and robustness across novel embodiments, objects, and sensor types (both taxel-based and vision-based tactile sensors). In real-world experiments, we demonstrate that our approach outperforms state-of-the-art visual trackers by a large margin. We further show that we can achieve precise manipulation tasks by incorporating our real-time object tracking result into motion plans, underscoring the advantages of visuo-haptic perception. Project website: https://ivl.cs.brown.edu/research/v-hop

cs.RO↗

arXiv2025

2D Gaussian Splatting with Semantic Alignment for Image Inpainting

Gaussian Splatting (GS), a recent technique for converting discrete points into continuous spatial representations, has shown promising results in 3D scene modeling and 2D image super-resolution. In this paper, we explore its untapped potential for image inpainting, which demands both locally coherent pixel synthesis and globally consistent semantic restoration. We propose the first image inpainting framework based on 2D Gaussian Splatting, which encodes incomplete images into a continuous field of 2D Gaussian splat coefficients and reconstructs the final image via a differentiable rasterization process. The continuous rendering paradigm of GS inherently promotes pixel-level coherence in the inpainted results. To improve efficiency and scalability, we introduce a patch-wise rasterization strategy that reduces memory overhead and accelerates inference. For global semantic consistency, we incorporate features from a pretrained DINO model. We observe that DINO's global features are naturally robust to small missing regions and can be effectively adapted to guide semantic alignment in large-mask scenarios, ensuring that the inpainted content remains contextually consistent with the surrounding scene. Extensive experiments on standard benchmarks demonstrate that our method achieves competitive performance in both quantitative metrics and perceptual quality, establishing a new direction for applying Gaussian Splatting to 2D image processing.

cs.CV↗

arXiv2025

Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey

Large reasoning models (LRMs) like OpenAI o1 and DeepSeek R1 have demonstrated impressive performance on complex reasoning tasks like mathematics and programming with long Chain-of-Thought (CoT) reasoning sequences (slow-thinking), compared with traditional large language models (fast-thinking). However, these reasoning models also face a huge challenge that generating unnecessarily lengthy and redundant reasoning chains even for trivial questions. This phenomenon leads to a significant waste of inference resources, increases the response time for simple queries, and hinders the practical application of LRMs in real-world products. To this end, it is crucial to shorten lengthy reasoning chains and learn adaptive reasoning between fast and slow thinking based on input difficulty. In this survey, we provide a comprehensive overview of recent progress in concise and adaptive thinking for efficient reasoning of LRMs, including methodologies, benchmarks, and challenges for future exploration. We hope this survey can help researchers quickly understand the landscape of this field and inspire novel adaptive thinking ideas to facilitate better usage of LRMs.

cs.AI↗

arXiv2025

UniTac: Whole-Robot Touch Sensing Without Tactile Sensors

Robots can better interact with humans and unstructured environments through touch sensing. However, most commercial robots are not equipped with tactile skins, making it challenging to achieve even basic touch-sensing functions, such as contact localization. We present UniTac, a data-driven whole-body touch-sensing approach that uses only proprioceptive joint sensors and does not require the installation of additional sensors. Our approach enables a robot equipped solely with joint sensors to localize contacts. Our goal is to democratize touch sensing and provide an off-the-shelf tool for HRI researchers to provide their robots with touch-sensing capabilities. We validate our approach on two platforms: the Franka robot arm and the Spot quadruped. On Franka, we can localize contact to within 8.0 centimeters, and on Spot, we can localize to within 7.2 centimeters at around 2,000 Hz on an RTX 3090 GPU without adding any additional sensors to the robot. Project website: https://ivl.cs.brown.edu/research/unitac.

cs.RO↗

arXiv2025

Revisiting Audio-Visual Segmentation with Vision-Centric Transformer

Audio-Visual Segmentation (AVS) aims to segment sound-producing objects in video frames based on the associated audio signal. Prevailing AVS methods typically adopt an audio-centric Transformer architecture, where object queries are derived from audio features. However, audio-centric Transformers suffer from two limitations: perception ambiguity caused by the mixed nature of audio, and weakened dense prediction ability due to visual detail loss. To address these limitations, we propose a new Vision-Centric Transformer (VCT) framework that leverages vision-derived queries to iteratively fetch corresponding audio and visual information, enabling queries to better distinguish between different sounding objects from mixed audio and accurately delineate their contours. Additionally, we also introduce a Prototype Prompted Query Generation (PPQG) module within our VCT framework to generate vision-derived queries that are both semantically aware and visually rich through audio prototype prompting and pixel context grouping, facilitating audio-visual information aggregation. Extensive experiments demonstrate that our VCT framework achieves new state-of-the-art performances on three subsets of the AVSBench dataset. The code is available at https://github.com/spyflying/VCT_AVS.

cs.CV↗

arXiv2025

Cooperative Integrated Sensing and Communication Networks: Analysis and Distributed Design

This paper proposes a cooperative integrated sensing and communication network (Co-ISACNet) adopting hybrid beamforming (HBF) architecture, which improves both radar sensing and communication performance. The main contributions of this work are four-fold. First, we introduce a novel cooperative sensing method for the considered Co-ISACNet, followed by a comprehensive analysis of this method. This analysis mathematically verifies the benefits of Co-ISACNet and provides insightful design guidelines. Second, to show the benefits of Co-ISACNet, we propose to jointly design the HBF to maximize the network communication capacity while satisfying the constraint of beampattern similarity for radar sensing, which results in a highly dimensional and non-convex problem. Third, to facilitate the joint design, we propose a novel distributed optimization framework based on proximal gradient and alternating direction method of multipliers, namely PANDA. Fourth, we further adopt the proposed PANDA framework to solve the joint HBF design problem for the Co-ISACNet. By using the proposed PANDA framework, all access points (APs) optimize the HBF in parallel, where each AP only requires local channel state information and limited message exchange among the APs. Such framework reduces significantly the computational complexity and thus has pronounced benefits in practical scenarios. Simulation results verify the effectiveness of the proposed algorithm compared with the conventional centralized algorithm and show the remarkable performance improvement of radar sensing and communication by deploying Co-ISACNet.

eess.SP↗

arXiv2025

Reinforcement Learning Tuning for VideoLLMs: Reward Design and Data Efficiency

Understanding real-world videos with complex semantics and long temporal dependencies remains a fundamental challenge in computer vision. Recent progress in multimodal large language models (MLLMs) has demonstrated strong capabilities in vision-language tasks, while reinforcement learning tuning (RLT) has further improved their reasoning abilities. In this work, we explore RLT as a post-training strategy to enhance the video-specific reasoning capabilities of MLLMs. Built upon the Group Relative Policy Optimization (GRPO) framework, we propose a dual-reward formulation that supervises both semantic and temporal reasoning through discrete and continuous reward signals. To facilitate effective preference-based optimization, we introduce a variance-aware data selection strategy based on repeated inference to identify samples that provide informative learning signals. We evaluate our approach across eight representative video understanding tasks, including VideoQA, Temporal Video Grounding, and Grounded VideoQA. Our method consistently outperforms supervised fine-tuning and existing RLT baselines, achieving superior performance with significantly less training data. These results underscore the importance of reward design and data selection in advancing reasoning-centric video understanding with MLLMs. Notably, The initial code release (two months ago) has now been expanded with updates, including optimized reward mechanisms and additional datasets. The latest version is available at https://github.com/appletea233/Temporal-R1 .

cs.CV↗

arXiv2025

LLaVA-ST: A Multimodal Large Language Model for Fine-Grained Spatial-Temporal Understanding

Recent advancements in multimodal large language models (MLLMs) have shown promising results, yet existing approaches struggle to effectively handle both temporal and spatial localization simultaneously. This challenge stems from two key issues: first, incorporating spatial-temporal localization introduces a vast number of coordinate combinations, complicating the alignment of linguistic and visual coordinate representations; second, encoding fine-grained temporal and spatial information during video feature compression is inherently difficult. To address these issues, we propose LLaVA-ST, a MLLM for fine-grained spatial-temporal multimodal understanding. In LLaVA-ST, we propose Language-Aligned Positional Embedding, which embeds the textual coordinate special token into the visual space, simplifying the alignment of fine-grained spatial-temporal correspondences. Additionally, we design the Spatial-Temporal Packer, which decouples the feature compression of temporal and spatial resolutions into two distinct point-to-region attention processing streams. Furthermore, we propose ST-Align dataset with 4.3M training samples for fine-grained spatial-temporal multimodal understanding. With ST-align, we present a progressive training pipeline that aligns the visual and textual feature through sequential coarse-to-fine stages.Additionally, we introduce an ST-Align benchmark to evaluate spatial-temporal interleaved fine-grained understanding tasks, which include Spatial-Temporal Video Grounding (STVG) , Event Localization and Captioning (ELC) and Spatial Video Grounding (SVG). LLaVA-ST achieves outstanding performance on 11 benchmarks requiring fine-grained temporal, spatial, or spatial-temporal interleaving multimodal understanding. Our code, data and benchmark will be released at Our code, data and benchmark will be released at https://github.com/appletea233/LLaVA-ST .

cs.CV↗

arXiv2025

Channel Estimation for Beyond Diagonal RIS via Tensor Decomposition

This paper addresses the channel estimation problem for beyond diagonal reconfigurable intelligent surface (BD-RIS) from a tensor decomposition perspective. We first show that the received pilot signals can be arranged as a three-way tensor, allowing us to recast the cascaded channel estimation problem as a block Tucker decomposition problem that yields decoupled estimates for the involved channel matrices while offering a substantial performance gain over the conventional (matrix-based) least squares (LS) estimation method. More specifically, we develop two solutions to solve the problem. The first one is a closed-form solution that extracts the channel estimates via a block Tucker Kronecker factorization (BTKF), which boils down to solving a set of parallel rank-one matrix approximation problems. Exploiting such a low-rank property yields a noise rejection gain compared to the standard LS estimation scheme while allowing the two involved channels to be estimated separately. The second solution is based on a block Tucker alternating least squares (BTALS) algorithm that directly estimates the involved channel matrices using an iterative estimation procedure. We discuss the uniqueness and identifiability issues and their implications for training design. We also propose a tensor-based design of the BD-RIS training tensor for each algorithm that ensures unique decoupled channel estimates under trivial scaling ambiguities. Our numerical results shed light on the tradeoffs offered by BTKF and BTALS methods. Specifically, while the first enjoys fast and parallel extraction of the channel estimates in closed form, the second has a more flexible training design, allowing for a significantly reduced training overhead compared to the state-of-the-art LS method.

eess.SP↗

arXiv2025

KMT-2022-BLG-1818Lb,c: A Cold Super-Jupiter with a Saturn Sibling

We present the discovery and analysis of the sixth microlensing two-planet system, KMT-2022-BLG-1818Lb,c, detected by a follow-up program targeting high-magnification events. Both planets are subject to the well-known ''Close/Wide'' degeneracy, although for the first planet, which has a super-Jovian mass ratio of $q_2 \simeq 5\times 10^{-3}$ in both solutions, the Close topology, with a normalized separation of $s\simeq 0.70$, is clearly preferred by $Δχ^2=26$. However, contrary to all previous two-planet microlensing systems, the mass ratio for the second planet, $q_3$, is substantially (factor of $\sim 10$) different for the Close and Wide topologies of the first planet. While this degeneracy is resolved in the present case due to high-cadence follow-up observations, the appearance of this new degeneracy indicates the need for caution in the analysis of future two-planet systems. A Bayesian analysis suggests that the host is likely a K-dwarf star in the Galactic disk. The first planet is probably a super-Jupiter on a Jupiter-like orbit, while the second planet is a Saturn-class planet on either a Mercury-like or Saturn-like orbit.

astro-ph.EP↗

arXiv2025

Non-Reciprocal Beyond Diagonal RIS: Multiport Network Models and Performance Benefits in Full-Duplex Systems

Beyond diagonal reconfigurable intelligent surfaces (BD-RIS) is a new advance in RIS techniques that introduces reconfigurable inter-element connections to generate scattering matrices not limited to being diagonal. BD-RIS has been recently proposed and proven to have benefits in enhancing channel gain and enlarging coverage in wireless communications. Uniquely, BD-RIS enables reciprocal and non-reciprocal architectures characterized by symmetric and non-symmetric scattering matrices. However, the performance benefits and new use cases enabled by non-reciprocal BD-RIS for wireless systems remain unexplored. This work takes a first step toward closing this knowledge gap and studies the non-reciprocal BD-RIS in full-duplex systems and its performance benefits over reciprocal counterparts. We start by deriving a general RIS aided full-duplex system model using a multiport circuit theory, followed by a simplified channel model based on physically consistent assumptions. With the considered channel model, we investigate the effect of BD-RIS non-reciprocity and identify the theoretical conditions for reciprocal and non-reciprocal BD-RISs to simultaneously achieve the maximum received power of the signal of interest in the uplink and the downlink. Simulation results validate the theories and highlight the significant benefits offered by non-reciprocal BD-RIS in full-duplex systems. The significant gains are achieved because of the non-reciprocity principle which implies that if a wave hits the non-reciprocal BD-RIS from one direction, the surface behaves differently than if it hits from the opposite direction. This enables an uplink user and a downlink user at different locations to optimally communicate with the same full-duplex base station via a non-reciprocal BD-RIS, which would not be possible with reciprocal surfaces.

eess.SP↗