arXiv subjects

Jingwen Zhao

Publications and source records attributed to Jingwen Zhao.

14 recordsLinked to original sources

arXiv2026

MVFA: A Multi-View Text-Guided Multimodal Fusion LLM Adapter for Sentiment Analysis and Emotion Recognition

Multimodal sentiment analysis and emotion recognition in conversations demand effective modeling of heterogeneous interactions across textual, acoustic, and visual modalities. Although large language models (LLMs) offer powerful language understanding, adapting them to multimodal affective computing remains challenging: full-model fine-tuning is computationally prohibitive, while many existing lightweight adapters fail to preserve rich textual cues during cross-modal fusion. To address these limitations, we propose the multi-view text-guided multimodal fusion adapter (MVFA), a parameter-efficient framework that augments frozen LLMs with strong multimodal reasoning capability. MVFA first constructs complementary text views via max pooling, mean pooling, and attention pooling; these views then guide cross-modal interactions with audio and visual features. The fused multimodal representations are subsequently compressed into a compact set of learnable pseudo-tokens through an Enhanced Q-Former Fusion Module. Using ChatGLM3-6B-base as the primary backbone, we further validate MVFA on LLaMA2-7B and Qwen3-8B to examine its portability across multiple frozen LLM backbones. MVFA is evaluated on three challenging datasets: CH-SIMS V2.0, MELD, and CHERMA. Experimental results demonstrate that MVFA achieves state-of-the-art performance on key metrics while updating only a small fraction of parameters. Specifically, it attains 84.62\% Acc2 and 84.59\% F1 on CH-SIMS V2.0, 67.36\% Acc and 66.03\% WF1 on MELD, and 74.66\% Acc on CHERMA. These findings establish multi-view text-guided fusion as an effective and scalable paradigm for parameter-efficient multimodal LLM adaptation in affective computing. The code is publicly available at https://github.com/Overwhelm1208/MVFA.

cs.AI↗

arXiv2026

On the number of factorable induced subgraphs

Let $F$ be an $r$-vertex graph. In this paper, we study the $F$-factor problem in random induced subgraphs of dense graphs. We show that for any $r$-vertex graph $F$ and $γ>0$, if $H$ is an $n$-vertex graph with minimum degree at least $(1-1/χ_{cr}(F)+γ)n$, then for every fixed $p \in (0,1)$, the random induced subgraph $H[p]$ contains an $F$-factor with probability at least $1/(rq)-o_n(1)$, where $q\in \mathbb{N}$ is the order of certain coset group defined from $H$. The probability is asymptotically best possible for infinitely many $F$ and $H$ and yields that a $1/(rq)-o_n(1)$ proportion of the subsets of $H$ induce $F$-factors, interestingly, regardless of whether $H$ itself admits an $F$-factor. Similar results are obtained for perfect matchings in hypergraphs under minimum degree conditions. Our proof combines concentration inequalities, lattice point counting in $\mathbb{Z}^d$ and structural theorems for $F$-factors in dense (hyper)graphs.

math.CO↗

arXiv2026

Perfect Matchings in Random Sparsifications of Dense Hypergraphs

Given $1\le\ell 1-(1-1/k)^{k-\ell}$ and this was recently verified by the work of Gan--Han, together with a very recent work of Fu et al. In this paper we study the existence of perfect matchings in the random $p$-sparsification of such $k$-uniform hypergraphs, that is, for $p=p(n)\in [0,1]$, each edge is selected independently with probability $p$. Building on the structural theory of Gan and Han, we show that the corresponding dense perfect matching results are robust under random sparsification. As consequences, we obtain deterministic polynomial-time algorithms that asymptotically almost surely solve the associated decision problems, as well as lower bounds on the number of perfect matchings in such hypergraphs -- interestingly, such hypergraphs either have no perfect matching, or have $(Ω(n))^{(1-1/k)n}$ perfect matchings. Moreover, we also establish analogous results for the $F$-factor problem in graphs. Our proofs combine a partial exposure algorithm, the lattice-based absorption method, and a random redistribution method of Kelly, Müyesser and Pokrovskiy, via the framework of spread distributions. A key new ingredient is a lattice-preparation step that separates the contributions of the two classes of robust index vectors arising in the Gan--Han structural theory. Together with the random redistribution method, this allows us to establish the desired spread property in the family of perfect matchings.

math.CO↗

arXiv2025

Robustness for expander graphs

We study robust versions of properties of $(n,d,λ)$-graphs, namely, the property of a random sparsification of an $(n,d,λ)$-graph, where each edge is retained with probability $p$ independently. We prove such results for the containment problem of perfect matchings, Hamiltonian cycles, and triangle factors. These results address a series of problems posed by Frieze and Krivelevich. First we prove that given $γ>0$, for sufficient large $n$, any $(n,d,λ)$-graph $G$ with $λ=o(d)$, $d=Ω(\log n)$ and $p\ge\frac{(1+γ)\log n}{d}$, $G\cap G(n,p)$ contains a Hamiltonian cycle (and thus a perfect matching if $n$ is even) with high probability. This result is asymptotically optimal. Moreover, we show that for sufficient large $n$, any $(n,d,λ)$-graph $G$ with $λ=o(\frac{d^2}{n})$, $d=Ω(n^{\frac{5}{6}}\log^{\frac{1}{2}}n)$ and $p\gg d^{-1}n^{\frac{1}{3}}\log^{\frac{1}{3}} n$, $G\cap G(n,p)$ contains a triangle factor with high probability. Here, the restrictions on $p$ and $λ$ are asymptotically optimal. Our proof for the triangle factor problem uses the iterative absorption approach to build a spread measure on the triangle factors, and we also prove and use a coupling result for triangles in the random subgraph of an expander $G$ and the hyperedges in the random subgraph of the triangle-hypergraph of $G$.

math.CO↗

arXiv2025

OneDB: A Distributed Multi-Metric Data Similarity Search System

Increasingly massive volumes of multi-modal data are being accumulated in many {real world} settings, including in health care and e-commerce. This development calls for effective general-purpose data management solutions for multi-modal data. Such a solution must facilitate user-friendly and accurate retrieval of any multi-modal data according to diverse application requirements. Further, such a solution must be capable of efficient and scalable retrieval. To address this need, we present OneDB, a distributed multi-metric data similarity retrieval system. This system exploits the fact that data of diverse modalities, such as text, images, and video, can be represented as metric data. The system thus affords each data modality its own metric space with its own distance function and then uses a multi-metric model to unify multi-modal data. The system features several innovations: (i) an extended Spart SQL query interface; (ii) lightweight means of learning appropriate weights of different modalities when retrieving multi-modal data to enable accurate retrieval; (iii) smart search-space pruning strategies that improve efficiency; (iv) two-layered indexing of data to ensure load-balancing during distributed processing; and (v) end-to-end system parameter autotuning. Experiments on three real-life datasets and two synthetic datasets offer evidence that the system is capable of state-of-the-art performance: (i) efficient and effective weight learning; (ii) retrieval accuracy improvements of 12.63\%--30.75\% over the state-of-the-art vector similarity search system at comparable efficiency; (iii) accelerated search by 2.5--5.75x over state-of-the-art single- or multi-metric solutions; (iv) demonstrated high scalability; and (v) parameter tuning that enables performance improvements of 15+%.

cs.DB↗

arXiv2025

Decoupled Seg Tokens Make Stronger Reasoning Video Segmenter and Grounder

Existing video segmenter and grounder approaches, exemplified by Sa2VA, directly fuse features within segmentation models. This often results in an undesirable entanglement of dynamic visual information and static semantics, thereby degrading segmentation accuracy. To systematically mitigate this issue, we propose DeSa2VA, a decoupling-enhanced prompting scheme integrating text pre-training and a linear decoupling module to address the information processing limitations inherent in SAM-2. Specifically, first, we devise a pre-training paradigm that converts textual ground-truth labels into point-level prompts while generating corresponding text masks. These masks are refined through a hybrid loss function to strengthen the model's semantic grounding capabilities. Next, we employ linear projection to disentangle hidden states that generated by a large language model into distinct textual and visual feature subspaces. Finally, a dynamic mask fusion strategy synergistically combines these decoupled features through triple supervision from predicted text/visual masks and ground-truth annotations. Extensive experiments demonstrate state-of-the-art performance across diverse tasks, including image segmentation, image question answering, video segmentation, and video question answering. Our codes are available at https://github.com/longmalongma/DeSa2VA.

cs.CV↗

arXiv2025

AIstorian lets AI be a historian: A KG-powered multi-agent system for accurate biography generation

Huawei has always been committed to exploring the AI application in historical research. Biography generation, as a specialized form of abstractive summarization, plays a crucial role in historical research but faces unique challenges that existing large language models (LLMs) struggle to address. These challenges include maintaining stylistic adherence to historical writing conventions, ensuring factual fidelity, and handling fragmented information across multiple documents. We present AIstorian, a novel end-to-end agentic system featured with a knowledge graph (KG)-powered retrieval-augmented generation (RAG) and anti-hallucination multi-agents. Specifically, AIstorian introduces an in-context learning based chunking strategy and a KG-based index for accurate and efficient reference retrieval. Meanwhile, AIstorian orchestrates multi-agents to conduct on-the-fly hallucination detection and error-type-aware correction. Additionally, to teach LLMs a certain language style, we finetune LLMs based on a two-step training approach combining data augmentation-enhanced supervised fine-tuning with stylistic preference optimization. Extensive experiments on a real-life historical Jinshi dataset demonstrate that AIstorian achieves a 3.8x improvement in factual accuracy and a 47.6% reduction in hallucination rate compared to existing baselines. The data and code are available at: https://github.com/ZJU-DAILY/AIstorian.

cs.CL↗

arXiv2025

Neuromorphic spatiotemporal optical flow: Enabling ultrafast visual perception beyond human capabilities

Optical flow, inspired by the mechanisms of biological visual systems, calculates spatial motion vectors within visual scenes that are necessary for enabling robotics to excel in complex and dynamic working environments. However, current optical flow algorithms, despite human-competitive task performance on benchmark datasets, remain constrained by unacceptable time delays (~0.6 seconds per inference, 4X human processing speed) in practical deployment. Here, we introduce a neuromorphic optical flow approach that addresses delay bottlenecks by encoding temporal information directly in a synaptic transistor array to assist spatial motion analysis. Compared to conventional spatial-only optical flow methods, our spatiotemporal neuromorphic optical flow offers the spatial-temporal consistency of motion information, rapidly identifying regions of interest in as little as 1-2 ms using the temporal motion cues derived from the embedded temporal information in the two-dimensional floating gate synaptic transistors. Thus, the visual input can be selectively filtered to achieve faster velocity calculations and various task execution. At the hardware level, due to the atomically sharp interfaces between distinct functional layers in two-dimensional van der Waals heterostructures, the synaptic transistor offers high-frequency response (~100 μs), robust non-volatility (>10000 s), and excellent endurance (>8000 cycles), enabling robust visual processing. In software benchmarks, our system outperforms state-of-the-art algorithms with a 400% speedup, frequently surpassing human-level performance while maintaining or enhancing accuracy by utilizing the temporal priors provided by the embedded temporal information.

cs.CV↗

arXiv2024

Hotspots and Trends in Magnetoencephalography Research (2013-2022): A Bibliometric Analysis

This study aimed to utilize bibliometric methods to analyze trends in international Magnetoencephalography (MEG) research from 2013 to 2022. Due to the limited volume of domestic literature on MEG, this analysis focuses solely on the global research landscape, providing insights from the past decade as a representative sample. This study utilized bibliometric methods to explore and analyze the progress, hotspots and developmental trends in international MEG research spanning from 1995 to 2022. The results indicated a dynamic and steady growth trend in the overall number of publications in MEG. Ryusuke Kakigi emerged as the most prolific author, while Neuroimage led as the most prolific journal. Current hotspots in MEG research encompass resting state, networks, functional connectivity, phase dynamics, oscillation, and more. Future trends in MEG research are poised to advance across three key aspects: disease treatment and practical applications, experimental foundations and technical advancements, and fundamental and advanced human cognition. In the future, there should be a focus on enhancing cross-integration and utilization of MEG with other instruments to diversify research methodologies in this field

cs.DL↗

arXiv2022

PSA-Det3D: Pillar Set Abstraction for 3D object Detection

Small object detection for 3D point cloud is a challenging problem because of two limitations: (1) Perceiving small objects is much more diffcult than normal objects due to the lack of valid points. (2) Small objects are easily blocked which breaks the shape of their meshes in 3D point cloud. In this paper, we propose a pillar set abstraction (PSA) and foreground point compensation (FPC) and design a point-based detection network, PSA-Det3D, to improve the detection performance for small object. The PSA embeds a pillar query operation on the basis of set abstraction (SA) to expand its receptive field of the network, which can aggregate point-wise features effectively. To locate more occluded objects, we persent a proposal generation layer consisting of a foreground point segmentation and a FPC module. Both the foreground points and the estimated centers are finally fused together to generate the detection result. The experiments on the KITTI 3D detection benchmark show that our proposed PSA-Det3D outperforms other algorithms with high accuracy for small object detection.

cs.CV↗

arXiv2022

MSE-Based Transceiver Designs for RIS-Aided Communications With Hardware Impairments

It is challenging to precisely configure the phase shifts of the reflecting elements at the reconfigurable intelligent surface (RIS) due to inherent hardware impairments (HIs). In this paper, the mean square error (MSE) performance is investigated in an RIS-aided single-user multiple-input multipleoutput (MIMO) communication system with transceiver HIs and RIS phase noise. We aim to jointly optimize the transmit precoder, linear received equalizer, and RIS reflecting matrices to minimize the MSE. To tackle this problem, an iterative algorithm is proposed, wherein the beamforming matrices are alternately optimized. Specifically, for the beamforming optimization subproblem, we derive the closed-form expression of the optimal precoder and equalizer matrices. Then, for the phase shift optimization subproblem, an efficient algorithm based on the majorization-minimization (MM) method is proposed. Simulation results show that the proposed MSE-based RIS-aided transceiver scheme dramatically outperforms the conventional system algorithms that do not consider HIs at both the transceiver and the RIS.

cs.IT↗

arXiv2021

Hierarchical Motion Encoder-Decoder Network for Trajectory Forecasting

Trajectory forecasting plays a pivotal role in the field of intelligent vehicles or social robots. Recent works focus on modeling spatial social impacts or temporal motion attentions, but neglect inherent properties of motions, i.e. moving trends and driving intentions. This paper proposes a context-free Hierarchical Motion Encoder-Decoder Network (HMNet) for vehicle trajectory prediction. HMNet first infers the hierarchical difference on motions to encode physically compliant patterns with high expressivity of moving trends and driving intentions. Then, a goal (endpoint)-embedded decoder hierarchically constructs multimodal predictions depending on the location-velocity-acceleration-related patterns. Besides, we present a modified social pooling module which considers certain motion properties to represent social interactions. HMNet enables to make the accurate, unimodal/multimodal and physically-socially-compliant prediction. Experiments on three public trajectory prediction datasets, i.e. NGSIM, HighD and Interaction show that our model achieves the state-of-the-art performance both quantitatively and qualitatively. We will release our code here: https://github.com/xuedashuai/HMNet.

cs.CV↗

arXiv2021

Energy Efficient Reconfigurable Intelligent Surface Enabled Mobile Edge Computing Networks with NOMA

Reconfigurable intelligent surface (RIS) has emerged as a promising technology for achieving high spectrum and energy efficiency in future wireless communication networks. In this paper, we investigate an RIS-aided single-cell multi-user mobile edge computing (MEC) system where an RIS is deployed to support the communication between a base station (BS) equipped with MEC servers and multiple single-antenna users. To utilize the scarce frequency resource efficiently, we assume that users communicate with BS based on a non-orthogonal multiple access (NOMA) protocol. Each user has a computation task which can be computed locally or partially/fully offloaded to the BS. We aim to minimize the sum energy consumption of all users by jointly optimizing the passive phase shifters, the size of transmission data, transmission rate, power control, transmission time and the decoding order. Since the resulting problem is non-convex, we use the block coordinate descent method to alternately optimize two separated subproblems. More specifically, we use the dual method to tackle a subproblem with given phase shift and obtain the closed-form solution; and then we utilize penalty method to solve another subproblem for given power control. Moreover, in order to demonstrate the effectiveness of our proposed algorithm, we propose three benchmark schemes: the time-division multiple access (TDMA)-MEC scheme, the full local computing scheme and the full offloading scheme. We use an alternating 1-D search method and the dual method that can solve the TDMA-based transmission problem well. Numerical results demonstrate that the proposed scheme can increase the energy efficiency and achieve significant performance gains over the three benchmark schemes.

cs.IT↗

arXiv2021

Spatial-Channel Transformer Network for Trajectory Prediction on the Traffic Scenes

Predicting motion of surrounding agents is critical to real-world applications of tactical path planning for autonomous driving. Due to the complex temporal dependencies and social interactions of agents, on-line trajectory prediction is a challenging task. With the development of attention mechanism in recent years, transformer model has been applied in natural language sequence processing first and then image processing. In this paper, we present a Spatial-Channel Transformer Network for trajectory prediction with attention functions. Instead of RNN models, we employ transformer model to capture the spatial-temporal features of agents. A channel-wise module is inserted to measure the social interaction between agents. We find that the Spatial-Channel Transformer Network achieves promising results on real-world trajectory prediction datasets on the traffic scenes.

cs.CV↗