arXiv subjects

Wen Wen

Publications and source records attributed to Wen Wen.

At least 37 records · Page 2Linked to original sources

arXiv2025

VQualA 2025 Challenge on Engagement Prediction for Short Videos: Methods and Results

This paper presents an overview of the VQualA 2025 Challenge on Engagement Prediction for Short Videos, held in conjunction with ICCV 2025. The challenge focuses on understanding and modeling the popularity of user-generated content (UGC) short videos on social media platforms. To support this goal, the challenge uses a new short-form UGC dataset featuring engagement metrics derived from real-world user interactions. This objective of the Challenge is to promote robust modeling strategies that capture the complex factors influencing user engagement. Participants explored a variety of multi-modal features, including visual content, audio, and metadata provided by creators. The challenge attracted 97 participants and received 15 valid test submissions, contributing significantly to progress in short-form UGC video engagement prediction.

cs.CV↗

arXiv2025

Engagement Prediction of Short Videos with Large Multimodal Models

The rapid proliferation of user-generated content (UGC) on short-form video platforms has made video engagement prediction increasingly important for optimizing recommendation systems and guiding content creation. However, this task remains challenging due to the complex interplay of factors such as semantic content, visual quality, audio characteristics, and user background. Prior studies have leveraged various types of features from different modalities, such as visual quality, semantic content, background sound, etc., but often struggle to effectively model their cross-feature and cross-modality interactions. In this work, we empirically investigate the potential of large multimodal models (LMMs) for video engagement prediction. We adopt two representative LMMs: VideoLLaMA2, which integrates audio, visual, and language modalities, and Qwen2.5-VL, which models only visual and language modalities. Specifically, VideoLLaMA2 jointly processes key video frames, text-based metadata, and background sound, while Qwen2.5-VL utilizes only key video frames and text-based metadata. Trained on the SnapUGC dataset, both models demonstrate competitive performance against state-of-the-art baselines, showcasing the effectiveness of LMMs in engagement prediction. Notably, VideoLLaMA2 consistently outperforms Qwen2.5-VL, highlighting the importance of audio features in engagement prediction. By ensembling two types of models, our method achieves first place in the ICCV VQualA 2025 EVQA-SnapUGC Challenge on short-form video engagement prediction. The code is available at https://github.com/sunwei925/LMM-EVQA.git.

cs.CV↗

arXiv2025

InfoSAM: Fine-Tuning the Segment Anything Model from An Information-Theoretic Perspective

The Segment Anything Model (SAM), a vision foundation model, exhibits impressive zero-shot capabilities in general tasks but struggles in specialized domains. Parameter-efficient fine-tuning (PEFT) is a promising approach to unleash the potential of SAM in novel scenarios. However, existing PEFT methods for SAM neglect the domain-invariant relations encoded in the pre-trained model. To bridge this gap, we propose InfoSAM, an information-theoretic approach that enhances SAM fine-tuning by distilling and preserving its pre-trained segmentation knowledge. Specifically, we formulate the knowledge transfer process as two novel mutual information-based objectives: (i) to compress the domain-invariant relation extracted from pre-trained SAM, excluding pseudo-invariant information as possible, and (ii) to maximize mutual information between the relational knowledge learned by the teacher (pre-trained SAM) and the student (fine-tuned model). The proposed InfoSAM establishes a robust distillation framework for PEFT of SAM. Extensive experiments across diverse benchmarks validate InfoSAM's effectiveness in improving SAM family's performance on real-world tasks, demonstrating its adaptability and superiority in handling specialized scenarios.

cs.CV↗

arXiv2025

Towards the Generalization of Multi-view Learning: An Information-theoretical Analysis

Multiview learning has drawn widespread attention for its efficacy in leveraging cross-view consensus and complementarity information to achieve a comprehensive representation of data. While multi-view learning has undergone vigorous development and achieved remarkable success, the theoretical understanding of its generalization behavior remains elusive. This paper aims to bridge this gap by developing information-theoretic generalization bounds for multi-view learning, with a particular focus on multi-view reconstruction and classification tasks. Our bounds underscore the importance of capturing both consensus and complementary information from multiple different views to achieve maximally disentangled representations. These results also indicate that applying the multi-view information bottleneck regularizer is beneficial for satisfactory generalization performance. Additionally, we derive novel data-dependent bounds under both leave-one-out and supersample settings, yielding computational tractable and tighter bounds. In the interpolating regime, we further establish the fast-rate bound for multi-view learning, exhibiting a faster convergence rate compared to conventional square-root bounds. Numerical results indicate a strong correlation between the true generalization gap and the derived bounds across various learning scenarios.

stat.ML↗

arXiv2024

Neural Directed Speech Enhancement with Dual Microphone Array in High Noise Scenario

In multi-speaker scenarios, leveraging spatial features is essential for enhancing target speech. While with limited microphone arrays, developing a compact multi-channel speech enhancement system remains challenging, especially in extremely low signal-to-noise ratio (SNR) conditions. To tackle this issue, we propose a triple-steering spatial selection method, a flexible framework that uses three steering vectors to guide enhancement and determine the enhancement range. Specifically, we introduce a causal-directed U-Net (CDUNet) model, which takes raw multi-channel speech and the desired enhancement width as inputs. This enables dynamic adjustment of steering vectors based on the target direction and fine-tuning of the enhancement region according to the angular separation between the target and interference signals. Our model with only a dual microphone array, excels in both speech quality and downstream task performance. It operates in real-time with minimal parameters, making it ideal for low-latency, on-device streaming applications.

eess.AS↗

arXiv2024

An Ensemble Approach to Short-form Video Quality Assessment Using Multimodal LLM

The rise of short-form videos, characterized by diverse content, editing styles, and artifacts, poses substantial challenges for learning-based blind video quality assessment (BVQA) models. Multimodal large language models (MLLMs), renowned for their superior generalization capabilities, present a promising solution. This paper focuses on effectively leveraging a pretrained MLLM for short-form video quality assessment, regarding the impacts of pre-processing and response variability, and insights on combining the MLLM with BVQA models. We first investigated how frame pre-processing and sampling techniques influence the MLLM's performance. Then, we introduced a lightweight learning-based ensemble method that adaptively integrates predictions from the MLLM and state-of-the-art BVQA models. Our results demonstrated superior generalization performance with the proposed ensemble approach. Furthermore, the analysis of content-aware ensemble weights highlighted that some video characteristics are not fully represented by existing BVQA models, revealing potential directions to improve BVQA models further.

cs.CV↗

arXiv2024

Role Play: Learning Adaptive Role-Specific Strategies in Multi-Agent Interactions

Zero-shot coordination problem in multi-agent reinforcement learning (MARL), which requires agents to adapt to unseen agents, has attracted increasing attention. Traditional approaches often rely on the Self-Play (SP) framework to generate a diverse set of policies in a policy pool, which serves to improve the generalization capability of the final agent. However, these frameworks may struggle to capture the full spectrum of potential strategies, especially in real-world scenarios that demand agents balance cooperation with competition. In such settings, agents need strategies that can adapt to varying and often conflicting goals. Drawing inspiration from Social Value Orientation (SVO)-where individuals maintain stable value orientations during interactions with others-we propose a novel framework called \emph{Role Play} (RP). RP employs role embeddings to transform the challenge of policy diversity into a more manageable diversity of roles. It trains a common policy with role embedding observations and employs a role predictor to estimate the joint role embeddings of other agents, helping the learning agent adapt to its assigned role. We theoretically prove that an approximate optimal policy can be achieved by optimizing the expected cumulative reward relative to an approximate role-based policy. Experimental results in both cooperative (Overcooked) and mixed-motive games (Harvest, CleanUp) reveal that RP consistently outperforms strong baselines when interacting with unseen agents, highlighting its robustness and adaptability in complex environments.

cs.MA↗

arXiv2024

Phi-3 Safety Post-Training: Aligning Language Models with a "Break-Fix" Cycle

Recent innovations in language model training have demonstrated that it is possible to create highly performant models that are small enough to run on a smartphone. As these models are deployed in an increasing number of domains, it is critical to ensure that they are aligned with human preferences and safety considerations. In this report, we present our methodology for safety aligning the Phi-3 series of language models. We utilized a "break-fix" cycle, performing multiple rounds of dataset curation, safety post-training, benchmarking, red teaming, and vulnerability identification to cover a variety of harm areas in both single and multi-turn scenarios. Our results indicate that this approach iteratively improved the performance of the Phi-3 models across a wide range of responsible AI benchmarks. Finally, we include additional red teaming strategies and evaluations that were used to test the safety behavior of Phi-3.5-mini and Phi-3.5-MoE, which were optimized for multilingual capabilities.

cs.CL↗

arXiv2024

Trembling Motion of Exciton-Polaritons Close to the Rashba-Dresselhaus Regime

We report the experimental emulation of trembling quantum motion, or Zitterbewegung, of exciton polaritons in a perovskite microcavity at room temperature. By introducing liquid crystal molecules into the microcavity, we achieve spinor states with synthetic Rashba-Dresselhaus spin-orbit coupling and tunable energy splitting. Under a resonant excitation, the polariton fluid exhibits clear trembling motion perpendicular to its flowing direction, accompanied by a unique spin pattern resembling interlocked fingers. Furthermore, leveraging on the sizable tunability of energy gaps by external electrical voltages, we observe the continuous transition of polariton Zitterbewegung from relativistic (small gaps) to non-relativistic (large gaps) regimes. Our findings pave the way for using exciton polaritons in the emulation of relativistic quantum physics.

cond-mat.mes-hall↗

arXiv2024

Observation of perovskite topological valley exciton-polaritons at room temperature

Topological exciton-polaritons are a burgeoning class of topological photonic systems distinguished by their hybrid nature as part-light, part-matter quasiparticles. Their further control over novel valley degree of freedom (DOF) has offered considerable potential for developing active topological optical devices towards information processing. However, the experimental demonstration of propagating topological exciton-polaritons with valley DOF remains elusive at room temperature. Here, employing a two-dimensional (2D) valley-Hall perovskite lattice, we report the experimental observation of valley-polarized topological exciton-polaritons and their valley-dependent propagations at room temperature. The 2D valley-Hall perovskite lattice consists of two mutually inverted honeycomb lattices with broken inversion symmetry. By measuring their band structure with angle-resolved photoluminescence spectra, we experimentally verify the existence of valley-polarized polaritonic topological kink states with a large gap opening of ~ 9 meV in the bearded interface at room temperature. Moreover, these valley-polarized states exhibit counter-propagating behaviors under a resonant excitation at room temperature. Our results not only expand the landscape of realizing topological exciton-polaritons, but also pave the way for the development of topological valleytronic devices employing exciton-polaritons with valley DOF at room temperature

physics.optics↗

arXiv2024

Learned Scanpaths Aid Blind Panoramic Video Quality Assessment

Panoramic videos have the advantage of providing an immersive and interactive viewing experience. Nevertheless, their spherical nature gives rise to various and uncertain user viewing behaviors, which poses significant challenges for panoramic video quality assessment (PVQA). In this work, we propose an end-to-end optimized, blind PVQA method with explicit modeling of user viewing patterns through visual scanpaths. Our method consists of two modules: a scanpath generator and a quality assessor. The scanpath generator is initially trained to predict future scanpaths by minimizing their expected code length and then jointly optimized with the quality assessor for quality prediction. Our blind PVQA method enables direct quality assessment of panoramic images by treating them as videos composed of identical frames. Experiments on three public panoramic image and video quality datasets, encompassing both synthetic and authentic distortions, validate the superiority of our blind PVQA model over existing methods.

eess.IV↗

arXiv2024

Analysis of Video Quality Datasets via Design of Minimalistic Video Quality Models

Blind video quality assessment (BVQA) plays an indispensable role in monitoring and improving the end-users' viewing experience in various real-world video-enabled media applications. As an experimental field, the improvements of BVQA models have been measured primarily on a few human-rated VQA datasets. Thus, it is crucial to gain a better understanding of existing VQA datasets in order to properly evaluate the current progress in BVQA. Towards this goal, we conduct a first-of-its-kind computational analysis of VQA datasets via designing minimalistic BVQA models. By minimalistic, we restrict our family of BVQA models to build only upon basic blocks: a video preprocessor (for aggressive spatiotemporal downsampling), a spatial quality analyzer, an optional temporal quality analyzer, and a quality regressor, all with the simplest possible instantiations. By comparing the quality prediction performance of different model variants on eight VQA datasets with realistic distortions, we find that nearly all datasets suffer from the easy dataset problem of varying severity, some of which even admit blind image quality assessment (BIQA) solutions. We additionally justify our claims by contrasting our model generalizability on these VQA datasets, and by ablating a dizzying set of BVQA design choices related to the basic building blocks. Our results cast doubt on the current progress in BVQA, and meanwhile shed light on good practices of constructing next-generation VQA datasets and models.

cs.CV↗

arXiv2024

Modular Blind Video Quality Assessment

Blind video quality assessment (BVQA) plays a pivotal role in evaluating and improving the viewing experience of end-users across a wide range of video-based platforms and services. Contemporary deep learning-based models primarily analyze video content in its aggressively subsampled format, while being blind to the impact of the actual spatial resolution and frame rate on video quality. In this paper, we propose a modular BVQA model and a method of training it to improve its modularity. Our model comprises a base quality predictor, a spatial rectifier, and a temporal rectifier, responding to the visual content and distortion, spatial resolution, and frame rate changes on video quality, respectively. During training, spatial and temporal rectifiers are dropped out with some probabilities to render the base quality predictor a standalone BVQA model, which should work better with the rectifiers. Extensive experiments on both professionally-generated content and user-generated content video databases show that our quality model achieves superior or comparable performance to current methods. Additionally, the modularity of our model offers an opportunity to analyze existing video quality databases in terms of their spatial and temporal complexity.

eess.IV↗

arXiv2024

Perceptual Quality Assessment of Virtual Reality Videos in the Wild

Investigating how people perceive virtual reality (VR) videos in the wild (i.e., those captured by everyday users) is a crucial and challenging task in VR-related applications due to complex authentic distortions localized in space and time. Existing panoramic video databases only consider synthetic distortions, assume fixed viewing conditions, and are limited in size. To overcome these shortcomings, we construct the VR Video Quality in the Wild (VRVQW) database, containing $502$ user-generated videos with diverse content and distortion characteristics. Based on VRVQW, we conduct a formal psychophysical experiment to record the scanpaths and perceived quality scores from $139$ participants under two different viewing conditions. We provide a thorough statistical analysis of the recorded data, observing significant impact of viewing conditions on both human scanpaths and perceived quality. Moreover, we develop an objective quality assessment model for VR videos based on pseudocylindrical representation and convolution. Results on the proposed VRVQW show that our method is superior to existing video quality assessment models. We have made the database and code available at https://github.com/limuhit/VR-Video-Quality-in-the-Wild.

cs.CV↗

arXiv2023

Vortex-lattice structures in rotating Bose-Fermi superfluid mixtures

The system of Bose-Fermi superfluid mixture offers a playground to explore rich macroscopic quantum phenomena. In a recent experiment of Yao {\it et al.} [Phys. Rev. Lett. {\bf 117}, 145301 (2016)], $^{41}$K-$^{6}$Li superfluid mixture is implemented. Coupled quantized vortices are generated via rotating the superfluid mixture, and a few unconventional behaviors on the formations of vortex numbers are observed, which can be traced to boson-fermion interactions. Here we provide a theoretical insight into the unconventional behaviors observed in the experiment. To this end, the orbital-free density functional theory is hired, and its utility is validated by making comparison of the numerical results and a microscopic theory for vortex lattices in strongly interacting Fermi superfluids alone. We also predict interesting phenomena which can be readily explored experimentally, including the novel structures of bosonic-fermionic vortices, and attractive interactions between the bosonic and fermionic vortices.

cond-mat.quant-gas↗

arXiv2023

Structure Prediction of Epitaxial Organic Interfaces with Ogre, Demonstrated for TCNQ on TTF

Highly ordered epitaxial interfaces between organic semiconductors are considered as a promising avenue for enhancing the performance of organic electronic devices including solar cells, light emitting diodes, and transistors, thanks to their well-controlled, uniform electronic properties and high carrier mobilities. Although the phenomenon of organic epitaxy has been known for decades, computational methods for structure prediction of epitaxial organic interfaces have lagged far behind the existing methods for their inorganic counterparts. We present a method for structure prediction of epitaxial organic interfaces based on lattice matching followed by surface matching, implemented in the open-source Python package, Ogre. The lattice matching step produces domain-matched interfaces, where commensurability is achieved with different integer multiples of the substrate and film unit cells. In the surface matching step, Bayesian optimization (BO) is used to find the interfacial distance and registry between the substrate and film. The BO objective function is based on dispersion corrected deep neural network interatomic potentials, shown to be in excellent agreement with density functional theory (DFT). The application of Ogre is demonstrated for an epitaxial interface of 7,7,8,8-tetracyanoquinodimethane (TCNQ) on tetrathiafulvalene (TTF), whose electronic structure has been probed by ultraviolet photoemission spectroscopy (UPS), but whose structure had been hitherto unknown [Organic Electronics 48, 371 (2017)]. We find that TCNQ(001) on top of TTF(100) is the most stable interface configuration, closely followed by TCNQ(010) on top of TTF(100). The density of states, calculated using DFT, is in excellent agreement with UPS, including the presence of an interface charge transfer state.

cond-mat.mtrl-sci↗

arXiv2021

Ensemble Making Few-Shot Learning Stronger

Few-shot learning has been proposed and rapidly emerging as a viable means for completing various tasks. Many few-shot models have been widely used for relation learning tasks. However, each of these models has a shortage of capturing a certain aspect of semantic features, for example, CNN on long-range dependencies part, Transformer on local features. It is difficult for a single model to adapt to various relation learning, which results in the high variance problem. Ensemble strategy could be competitive on improving the accuracy of few-shot relation extraction and mitigating high variance risks. This paper explores an ensemble approach to reduce the variance and introduces fine-tuning and feature attention strategies to calibrate relation-level features. Results on several few-shot relation learning tasks show that our model significantly outperforms the previous state-of-the-art models.

cs.CL↗

arXiv2020

Semi-Supervised Disentangled Framework for Transferable Named Entity Recognition

Named entity recognition (NER) for identifying proper nouns in unstructured text is one of the most important and fundamental tasks in natural language processing. However, despite the widespread use of NER models, they still require a large-scale labeled data set, which incurs a heavy burden due to manual annotation. Domain adaptation is one of the most promising solutions to this problem, where rich labeled data from the relevant source domain are utilized to strengthen the generalizability of a model based on the target domain. However, the mainstream cross-domain NER models are still affected by the following two challenges (1) Extracting domain-invariant information such as syntactic information for cross-domain transfer. (2) Integrating domain-specific information such as semantic information into the model to improve the performance of NER. In this study, we present a semi-supervised framework for transferable NER, which disentangles the domain-invariant latent variables and domain-specific latent variables. In the proposed framework, the domain-specific information is integrated with the domain-specific latent variables by using a domain predictor. The domain-specific and domain-invariant latent variables are disentangled using three mutual information regularization terms, i.e., maximizing the mutual information between the domain-specific latent variables and the original embedding, maximizing the mutual information between the domain-invariant latent variables and the original embedding, and minimizing the mutual information between the domain-specific and domain-invariant latent variables. Extensive experiments demonstrated that our model can obtain state-of-the-art performance with cross-domain and cross-lingual NER benchmark data sets.

cs.CL↗