Wei‐Yao Wang

Hao Tang, Kevin Liang, Matt Feiszli et al.

Visual object tracking is a key component to many egocentric vision problems. However, the full spectrum of challenges of egocentric tracking faced by an embodied AI is underrepresented in many existing datasets; these tend to focus on relatively short, third-person videos. Egocentric video has several distinguishing characteristics from those commonly found in past datasets: frequent large camera motions and hand interactions with objects commonly lead to occlusions or objects exiting the frame, and object appearance can change rapidly due to widely different points of view, scale, or object states. Embodied tracking is also naturally long-term, and being able to consistently (re-)associate objects to their appearances and disappearances over as long as a lifetime is critical. Previous datasets under-emphasize this re-detection problem, and their "framed" nature has led to adoption of various spatiotemporal priors that we find do not necessarily generalize to egocentric video. We thus introduce EgoTracks, a new dataset for long-term egocentric visual object tracking. Sourced from the Ego4D dataset, this new dataset presents a significant challenge to recent state-of-the-art single-object tracking models, which we find score poorly on traditional tracking metrics for our new dataset, compared to popular benchmarks. We further show improvements that can be made to a STARK tracker to significantly increase its performance on egocentric data, resulting in a baseline model we call EgoSTARK. We publicly release our annotations and benchmark, hoping our dataset leads to further advancements in tracking.

Wei-Wei Du, Hong-Wei Wu, Wei-Yao Wang et al.

Multi-modal fact verification has become an important but challenging issue on social media due to the mismatch between the text and images in the misinformation of news content, which has been addressed by considering cross-modalities to identify the veracity of the news in recent years. In this paper, we propose the Pre-CoFactv2 framework with new parameter-efficient foundation models for modeling fine-grained text and input embeddings with lightening parameters, multi-modal multi-type fusion for not only capturing relations for the same and different modalities but also for different types (i.e., claim and document), and feature representations for explicitly providing metadata for each sample. In addition, we introduce a unified ensemble method to boost model performance by adjusting the importance of each trained model with not only the weights but also the powers. Extensive experiments show that Pre-CoFactv2 outperforms Pre-CoFact by a large margin and achieved new state-of-the-art results at the Factify challenge at AAAI 2023. We further illustrate model variations to verify the relative contributions of different components. Our team won the first prize (F1-score: 81.82%) and we made our code publicly available at https://github.com/wwweiwei/Pre-CoFactv2-AAAI-2023.

Wei-Yao Wang, Wei-Wei Du, Wen-Chih Peng et al.

In recent years, badminton analytics has drawn attention due to the advancement of artificial intelligence and the efficiency of data collection. While there is a line of effective applications to improve and investigate player performance, there are only a few public badminton datasets that can be used by researchers outside the badminton domain. Existing badminton singles datasets focus on specific matchups; however, they cannot provide comprehensive studies on different players and various matchups. In this paper, we provide a badminton singles dataset, ShuttleSet22, which is collected from high-ranking matches in 2022. ShuttleSet22 consists of 30,172 strokes in 2,888 rallies in the training set, 1,400 strokes in 450 rallies in the validation set, and 2,040 strokes in 654 rallies in the testing set, with detailed stroke-level metadata within a rally. To benchmark existing work with ShuttleSet22, we hold a challenge, Track 2: Forecasting Future Turn-Based Strokes in Badminton Rallies, at CoachAI Badminton Challenge @ IJCAI 2023, to encourage researchers to tackle this real-world problem through innovative approaches and to summarize insights between the state-of-the-art baseline and improved techniques, exchanging inspiring ideas. The baseline codes and the dataset are made available at https://github.com/wywyWang/CoachAI-Projects/tree/main/CoachAI-Challenge-IJCAI2023.

Tarun Kalluri, Weiyao Wang, Heng Wang et al.

Many top-down architectures for instance segmentation achieve significant success when trained and tested on pre-defined closed-world taxonomy. However, when deployed in the open world, they exhibit notable bias towards seen classes and suffer from significant performance drop. In this work, we propose a novel approach for open world instance segmentation called bottom-Up and top-Down Open-world Segmentation (UDOS) that combines classical bottom-up segmentation algorithms within a top-down learning framework. UDOS first predicts parts of objects using a top-down network trained with weak supervision from bottom-up segmentations. The bottom-up segmentations are class-agnostic and do not overfit to specific taxonomies. The part-masks are then fed into affinity-based grouping and refinement modules to predict robust instance-level segmentations. UDOS enjoys both the speed and efficiency from the top-down architectures and the generalization ability to unseen categories from bottom-up supervision. We validate the strengths of UDOS on multiple cross-category as well as cross-dataset transfer tasks from 5 challenging datasets including MS-COCO, LVIS, ADE20k, UVO and OpenImages, achieving significant improvements over state-of-the-art across the board. Our code and models are available on our project page.

Pierre Gleize, Weiyao Wang, Matt Feiszli

Keypoint detection & descriptors are foundational tech-nologies for computer vision tasks like image matching, 3D reconstruction and visual odometry. Hand-engineered methods like Harris corners, SIFT, and HOG descriptors have been used for decades; more recently, there has been a trend to introduce learning in an attempt to improve keypoint detectors. On inspection however, the results are difficult to interpret; recent learning-based methods employ a vast diversity of experimental setups and design choices: empirical results are often reported using different backbones, protocols, datasets, types of supervisions or tasks. Since these differences are often coupled together, it raises a natural question on what makes a good learned keypoint detector. In this work, we revisit the design of existing keypoint detectors by deconstructing their methodologies and identifying the key components. We re-design each component from first-principle and propose Simple Learned Keypoints (SiLK) that is fully-differentiable, lightweight, and flexible. Despite its simplicity, SiLK advances new state-of-the-art on Detection Repeatability and Homography Estimation tasks on HPatches and 3D Point-Cloud Registration task on ScanNet, and achieves competitive performance to state-of-the-art on camera pose estimation in 2022 Image Matching Challenge and ScanNet.

Xiusi Chen, Wei-Yao Wang, Ziniu Hu et al.

Dynamically planning in complex systems has been explored to improve decision-making in various domains. Professional basketball serves as a compelling example of a dynamic spatio-temporal game, encompassing context-dependent decision-making. However, processing the diverse on-court signals and navigating the vast space of potential actions and outcomes make it difficult for existing approaches to swiftly identify optimal strategies in response to evolving circumstances. In this study, we formulate the sequential decision-making process as a conditional trajectory generation process. Based on the formulation, we introduce PlayBest (PLAYer BEhavior SynThesis), a method to improve player decision-making. We extend the diffusion probabilistic model to learn challenging environmental dynamics from historical National Basketball Association (NBA) player motion tracking data. To incorporate data-driven strategies, an auxiliary value function is trained with corresponding rewards. To accomplish reward-guided trajectory generation, we condition the diffusion model on the value function via classifier-guided sampling. We validate the effectiveness of PlayBest through simulation studies, contrasting the generated trajectories with those employed by professional basketball teams. Our results reveal that the model excels at generating reasonable basketball trajectories that produce efficient plays. Moreover, the synthesized play strategies exhibit an alignment with professional tactics, highlighting the model's capacity to capture the intricate dynamics of basketball games.

Anh Thai, Weiyao Wang, Hao Tang et al.

3D object part segmentation is essential in computer vision applications. While substantial progress has been made in 2D object part segmentation, the 3D counterpart has received less attention, in part due to the scarcity of annotated 3D datasets, which are expensive to collect. In this work, we propose to leverage a few annotated 3D shapes or richly annotated 2D datasets to perform 3D object part segmentation. We present our novel approach, termed 3-By-2 that achieves SOTA performance on different benchmarks with various granularity levels. By using features from pretrained foundation models and exploiting semantic and geometric correspondences, we are able to overcome the challenges of limited 3D annotations. Our approach leverages available 2D labels, enabling effective 3D object part segmentation. Our method 3-By-2 can accommodate various part taxonomies and granularities, demonstrating interesting part label transfer ability across different object categories. Project website: \url{https://ngailapdi.github.io/projects/3by2/}.

Wei-Yao Wang, Yung-Chang Huang, Tsi-Ui Ik et al.

With the recent progress in sports analytics, deep learning approaches have demonstrated the effectiveness of mining insights into players' tactics for improving performance quality and fan engagement. This is attributed to the availability of public ground-truth datasets. While there are a few available datasets for turn-based sports for action detection, these datasets severely lack structured source data and stroke-level records since these require high-cost labeling efforts from domain experts and are hard to detect using automatic techniques. Consequently, the development of artificial intelligence approaches is significantly hindered when existing models are applied to more challenging structured turn-based sequences. In this paper, we present ShuttleSet, the largest publicly-available badminton singles dataset with annotated stroke-level records. It contains 104 sets, 3,685 rallies, and 36,492 strokes in 44 matches between 2018 and 2021 with 27 top-ranking men's singles and women's singles players. ShuttleSet is manually annotated with a computer-aided labeling tool to increase the labeling efficiency and effectiveness of selecting the shot type with a choice of 18 distinct classes, the corresponding hitting locations, and the locations of both players at each stroke. In the experiments, we provide multiple benchmarks (i.e., stroke influence, stroke forecasting, and movement forecasting) with baselines to illustrate the practicability of using ShuttleSet for turn-based analytics, which is expected to stimulate both academic and sports communities. Over the past two years, a visualization platform has been deployed to illustrate the variability of analysis cases from ShuttleSet for coaches to delve into players' tactical preferences with human-interactive interfaces, which was also used by national badminton teams during multiple international high-ranking matches.

Kai-Shiang Chang, Wei-Yao Wang, Wen-Chih Peng

Sports analytics has captured increasing attention since analysis of the various data enables insights for training strategies, player evaluation, etc. In this paper, we focus on predicting what types of returning strokes will be made, and where players will move to based on previous strokes. As this problem has not been addressed to date, movement forecasting can be tackled through sequence-based and graph-based models by formulating as a sequence prediction task. However, existing sequence-based models neglect the effects of interactions between players, and graph-based models still suffer from multifaceted perspectives on the next movement. Moreover, there is no existing work on representing strategic relations among players' shot types and movements. To address these challenges, we first introduce the procedure of the Player Movements (PM) graph to exploit the structural movements of players with strategic relations. Based on the PM graph, we propose a novel Dynamic Graphs and Hierarchical Fusion for Movement Forecasting model (DyMF) with interaction style extractors to capture the mutual interactions of players themselves and between both players within a rally, and dynamic players' tactics across time. In addition, hierarchical fusion modules are designed to incorporate the style influence of both players and rally interactions. Extensive experiments show that our model empirically outperforms both sequence- and graph-based methods and demonstrate the practical usage of movement forecasting.

Chih-Chia Li, Wei-Yao Wang, Wei-Wei Du et al.

Real estate appraisal is a crucial issue for urban applications, which aims to value the properties on the market. Traditional methods perform appraisal based on the domain knowledge, but suffer from the efforts of hand-crafted design. Recently, several methods have been developed to automatize the valuation process by taking the property trading transaction into account when estimating the property value. However, existing methods only consider the real estate itself, ignoring the relation between the properties. Moreover, naively aggregating the information of neighbors fails to model the relationships between the transactions. To tackle these limitations, we propose a novel Neighbor Relation Graph Learning Framework (ReGram) by incorporating the relation between target transaction and surrounding neighbors with the attention mechanism. To model the influence between communities, we integrate the environmental information and the past price of each transaction from other communities. Moreover, since the target transactions in different regions share some similarities and differences of characteristics, we introduce a dynamic adapter to model the different distributions of the target transactions based on the input-related kernel weights. Extensive experiments on the real-world dataset with various scenarios demonstrate that ReGram robustly outperforms the state-of-the-art methods. Furthermore, comprehensive ablation studies were conducted to examine the effectiveness of each component in ReGram.

Li-Chun Huang, Nai-Zen Hseuh, Yen-Che Chien et al.

Recent techniques for analyzing sports precisely has stimulated various approaches to improve player performance and fan engagement. However, existing approaches are only able to evaluate offline performance since testing in real-time matches requires exhaustive costs and cannot be replicated. To test in a safe and reproducible simulator, we focus on turn-based sports and introduce a badminton environment by simulating rallies with different angles of view and designing the states, actions, and training procedures. This benefits not only coaches and players by simulating past matches for tactic investigation, but also researchers from rapidly evaluating their novel algorithms.

Hao Tang, Weiyao Wang, Pierre Gleize et al.

Recovering camera poses from a set of images is a foundational task in 3D computer vision, which powers key applications such as 3D scene/object reconstructions. Classic methods often depend on feature correspondence, such as keypoints, which require the input images to have large overlap and small viewpoint changes. Such requirements present considerable challenges in scenarios with sparse views. Recent data-driven approaches aim to directly output camera poses, either through regressing the 6DoF camera poses or formulating rotation as a probability distribution. However, each approach has its limitations. On one hand, directly regressing the camera poses can be ill-posed, since it assumes a single mode, which is not true under symmetry and leads to sub-optimal solutions. On the other hand, probabilistic approaches are capable of modeling the symmetry ambiguity, yet they sample the entire space of rotation uniformly by brute-force. This leads to an inevitable trade-off between high sample density, which improves model precision, and sample efficiency that determines the runtime. In this paper, we propose ADen to unify the two frameworks by employing a generator and a discriminator: the generator is trained to output multiple hypotheses of 6DoF camera pose to represent a distribution and handle multi-mode ambiguity, and the discriminator is trained to identify the hypothesis that best explains the data. This allows ADen to combine the best of both worlds, achieving substantially higher precision as well as lower runtime than previous methods in empirical evaluations.

Cheng-Ming Lin, Ching Chang, Wei-Yao Wang et al.

In recent years, microservices have gained widespread adoption in IT operations due to their scalability, maintenance, and flexibility. However, it becomes challenging for site reliability engineers (SREs) to pinpoint the root cause due to the complex relationships in microservices when facing system malfunctions. Previous research employed structured learning methods (e.g., PC-algorithm) to establish causal relationships and derive root causes from causal graphs. Nevertheless, they ignored the temporal order of time series data and failed to leverage the rich information inherent in the temporal relationships. For instance, in cases where there is a sudden spike in CPU utilization, it can lead to an increase in latency for other microservices. However, in this scenario, the anomaly in CPU utilization occurs before the latency increase, rather than simultaneously. As a result, the PC-algorithm fails to capture such characteristics. To address these challenges, we propose RUN, a novel approach for root cause analysis using neural Granger causal discovery with contrastive learning. RUN enhances the backbone encoder by integrating contextual information from time series, and leverages a time series forecasting model to conduct neural Granger causal discovery. In addition, RUN incorporates Pagerank with a personalization vector to efficiently recommend the top-k root causes. Extensive experiments conducted on the synthetic and real-world microservice-based datasets demonstrate that RUN noticeably outperforms the state-of-the-art root cause analysis methods. Moreover, we provide an analysis scenario for the sock-shop case to showcase the practicality and efficacy of RUN in microservice-based applications. Our code is publicly available at https://github.com/zmlin1998/RUN.

Yu-Chien Tang, Wei-Yao Wang, An-Zi Yen et al.

The dialogue systems in customer services have been developed with neural models to provide users with precise answers and round-the-clock support in task-oriented conversations by detecting customer intents based on their utterances. Existing intent detection approaches have highly relied on adaptively pre-training language models with large-scale datasets, yet the predominant cost of data collection may hinder their superiority. In addition, they neglect the information within the conversational responses of the agents, which have a lower collection cost, but are significant to customer intent as agents must tailor their replies based on the customers' intent. In this paper, we propose RSVP, a self-supervised framework dedicated to task-oriented dialogues, which utilizes agent responses for pre-training in a two-stage manner. Specifically, we introduce two pre-training tasks to incorporate the relations of utterance-response pairs: 1) Response Retrieval by selecting a correct response from a batch of candidates, and 2) Response Generation by mimicking agents to generate the response to a given utterance. Our benchmark results for two real-world customer service datasets show that RSVP significantly outperforms the state-of-the-art baselines by 4.95% for accuracy, 3.4% for MRR@3, and 2.75% for MRR@5 on average. Extensive case studies are investigated to show the validity of incorporating agent responses into the pre-training stage.

Weiyao Wang, Byung-Hak Kim, Varun Ganapathi

Recent advances in self-supervised learning (SSL) using large models to learn visual representations from natural images are rapidly closing the gap between the results produced by fully supervised learning and those produced by SSL on downstream vision tasks. Inspired by this advancement and primarily motivated by the emergence of tabular and structured document image applications, we investigate which self-supervised pretraining objectives, architectures, and fine-tuning strategies are most effective. To address these questions, we introduce RegCLR, a new self-supervised framework that combines contrastive and regularized methods and is compatible with the standard Vision Transformer architecture. Then, RegCLR is instantiated by integrating masked autoencoders as a representative example of a contrastive method and enhanced Barlow Twins as a representative example of a regularized method with configurable input image augmentations in both branches. Several real-world table recognition scenarios (e.g., extracting tables from document images), ranging from standard Word and Latex documents to even more challenging electronic health records (EHR) computer screen images, have been shown to benefit greatly from the representations learned from this new framework, with detection average-precision (AP) improving relatively by 4.8% for Table, 11.8% for Column, and 11.1% for GUI objects over a previous fully supervised baseline on real-world EHR screen images.

Ching Chang, Chiao-Tung Chan, Wei-Yao Wang et al.

Multivariate time-series data in numerous real-world applications (e.g., healthcare and industry) are informative but challenging due to the lack of labels and high dimensionality. Recent studies in self-supervised learning have shown their potential in learning rich representations without relying on labels, yet they fall short in learning disentangled embeddings and addressing issues of inductive bias (e.g., transformation-invariance). To tackle these challenges, we propose TimeDRL, a generic multivariate time-series representation learning framework with disentangled dual-level embeddings. TimeDRL is characterized by three novel features: (i) disentangled derivation of timestamp-level and instance-level embeddings from patched time-series data using a [CLS] token strategy; (ii) utilization of timestamp-predictive and instance-contrastive tasks for disentangled representation learning, with the former optimizing timestamp-level embeddings with predictive loss, and the latter optimizing instance-level embeddings with contrastive loss; and (iii) avoidance of augmentation methods to eliminate inductive biases, such as transformation-invariance from cropping and masking. Comprehensive experiments on 6 time-series forecasting datasets and 5 time-series classification datasets have shown that TimeDRL consistently surpasses existing representation learning approaches, achieving an average improvement of forecasting by 58.02% in MSE and classification by 1.48% in accuracy. Furthermore, extensive ablation studies confirmed the relative contribution of each component in TimeDRL's architecture, and semi-supervised learning evaluations demonstrated its effectiveness in real-world scenarios, even with limited labeled data. The code is available at https://github.com/blacksnail789521/TimeDRL.

Ying-Ying Chang, Wei-Yao Wang, Wen-Chih Peng

In the dynamic and rapidly evolving world of social media, detecting anomalous users has become a crucial task to address malicious activities such as misinformation and cyberbullying. As the increasing number of anomalous users improves the ability to mimic normal users and evade detection, existing methods only focusing on bot detection are ineffective in terms of capturing subtle distinctions between users. To address these challenges, we proposed SeGA, preference-aware self-contrastive learning for anomalous user detection, which leverages heterogeneous entities and their relations in the Twittersphere to detect anomalous users with different malicious strategies. SeGA utilizes the knowledge of large language models to summarize user preferences via posts. In addition, integrating user preferences with prompts as pseudo-labels for preference-aware self-contrastive learning enables the model to learn multifaceted aspects for describing the behaviors of users. Extensive experiments on the proposed TwBNT benchmark demonstrate that SeGA significantly outperforms the state-of-the-art methods (+3.5\% ~ 27.6\%) and empirically validate the effectiveness of the model design and pre-training strategies. Our code and data are publicly available at https://github.com/ying0409/SeGA.

Zhao Wang, Sota Moriyama, Wei-Yao Wang et al.

Recent advancements in LLM-based multi-agent (LLM-MA) systems have shown promise, yet significant challenges remain in managing communication and refinement when agents collaborate on complex tasks. In this paper, we propose \textit{Talk Structurally, Act Hierarchically (TalkHier)}, a novel framework that introduces a structured communication protocol for context-rich exchanges and a hierarchical refinement system to address issues such as incorrect outputs, falsehoods, and biases. \textit{TalkHier} surpasses various types of SoTA, including inference scaling model (OpenAI-o1), open-source multi-agent models (e.g., AgentVerse), and majority voting strategies on current LLM and single-agent baselines (e.g., ReAct, GPT4o), across diverse tasks, including open-domain question answering, domain-specific selective questioning, and practical advertisement text generation. These results highlight its potential to set a new standard for LLM-MA systems, paving the way for more effective, adaptable, and collaborative multi-agent frameworks. The code is available https://github.com/sony/talkhier.

Wei-Yao Wang, Zhao Wang, Helen Suzuki et al.

Recent Multimodal Large Language Models (MLLMs) have demonstrated significant progress in perceiving and reasoning over multimodal inquiries, ushering in a new research era for foundation models. However, vision-language misalignment in MLLMs has emerged as a critical challenge, where the textual responses generated by these models are not factually aligned with the given text-image inputs. Existing efforts to address vision-language misalignment have focused on developing specialized vision-language connectors or leveraging visual instruction tuning from diverse domains. In this paper, we tackle this issue from a fundamental yet unexplored perspective by revisiting the core architecture of MLLMs. Most MLLMs are typically built on decoder-only LLMs consisting of a causal attention mechanism, which limits the ability of the earlier modalities (e.g., images) to incorporate information from the latter modalities (e.g., text). To address this problem, we propose \MapleLeaf AKI, a novel MLLM that unlocks causal attention into modality-mutual attention (MMA) to enable image tokens to attend to text tokens. This simple yet effective design allows AKI to achieve superior performance in 12 multimodal understanding benchmarks (+7.2% on average) without introducing additional parameters and increasing training time. Our MMA design is intended to be generic, allowing for application across various modalities, and scalable to accommodate diverse multimodal scenarios. The code and model are publicly available at https://github.com/sony/aki to encourage further advancements in MLLMs across various directions.

Wei-Wei Du, Wei-Yao Wang, Wen-Chih Peng

Recommendation systems have illustrated the significant progress made in characterizing users' preferences based on their past behaviors. Despite the effectiveness of recommending accurately, there exist several factors that are essential but unexplored for evaluating various facets of recommendation systems, e.g., fairness, diversity, and limited resources. To address these issues, we propose Track2Vec, a GPU-free customizable-driven framework for fairness music recommendation. In order to take both accuracy and fairness into account, our solution consists of three modules, a customized fairness-aware groups for modeling different features based on configurable settings, a track representation learning module for learning better user embedding, and an ensemble module for ranking the recommendation results from different track representation learning modules. Moreover, inspired by TF-IDF which has been widely used in natural language processing, we introduce a metric called Miss Rate - Inverse Ground Truth Frequency (MR-ITF) to measure the fairness. Extensive experiments demonstrate that our model achieves a 4th price ranking in a GPU-free environment on the leaderboard in the EvalRS @ CIKM 2022 challenge, which is superior to the official baseline by about 200% in terms of the official scores. In addition, the ablation study illustrates the necessity of ensembling each group to acquire both accurate and fair recommendations.

Hong-Wei Wu, Wei-Yao Wang, Kuang-Da Wang et al.

Tabular data are fundamental in common machine learning applications, ranging from finance to genomics and healthcare. This paper focuses on tabular regression tasks, a field where deep learning (DL) methods are not consistently superior to machine learning (ML) models due to the challenges posed by irregular target functions inherent in tabular data, causing sensitive label changes with minor variations from features. To address these issues, we propose a novel Arithmetic-Aware Pre-training and Adaptive-Regularized Fine-tuning framework (APAR), which enables the model to fit irregular target function in tabular data while reducing the negative impact of overfitting. In the pre-training phase, APAR introduces an arithmetic-aware pretext objective to capture intricate sample-wise relationships from the perspective of continuous labels. In the fine-tuning phase, a consistency-based adaptive regularization technique is proposed to self-learn appropriate data augmentation. Extensive experiments across 10 datasets demonstrated that APAR outperforms existing GBDT-, supervised NN-, and pretrain-finetune NN-based methods in RMSE (+9.43% $\sim$ 20.37%), and empirically validated the effects of pre-training tasks, including the study of arithmetic operations. Our code and data are publicly available at https://github.com/johnnyhwu/APAR.

SAM 3D Team, Xingyu Chen, Fu-Jen Chu et al.

We present SAM 3D, a generative model for visually grounded 3D object reconstruction, predicting geometry, texture, and layout from a single image. SAM 3D excels in natural images, where occlusion and scene clutter are common and visual recognition cues from context play a larger role. We achieve this with a human- and model-in-the-loop pipeline for annotating object shape, texture, and pose, providing visually grounded 3D reconstruction data at unprecedented scale. We learn from this data in a modern, multi-stage training framework that combines synthetic pretraining with real-world alignment, breaking the 3D "data barrier". We obtain significant gains over recent work, with at least a 5:1 win rate in human preference tests on real-world objects and scenes. We will release our code and model weights, an online demo, and a new challenging benchmark for in-the-wild 3D object reconstruction.

Wei-Wei Du, Wei-Yao Wang, Wen-Chih Peng

The marketplace system connecting demands and supplies has been explored to develop unbiased decision-making in valuing properties. Real estate appraisal serves as one of the high-cost property valuation tasks for financial institutions since it requires domain experts to appraise the estimation based on the corresponding knowledge and the judgment of the market. Existing automated valuation models reducing the subjectivity of domain experts require a large number of transactions for effective evaluation, which is predominantly limited to not only the labeling efforts of transactions but also the generalizability of new developing and rural areas. To learn representations from unlabeled real estate sets, existing self-supervised learning (SSL) for tabular data neglects various important features, and fails to incorporate domain knowledge. In this paper, we propose DoRA, a Domain-based self-supervised learning framework for low-resource Real estate Appraisal. DoRA is pre-trained with an intra-sample geographic prediction as the pretext task based on the metadata of the real estate for equipping the real estate representations with prior domain knowledge. Furthermore, inter-sample contrastive learning is employed to generalize the representations to be robust for limited transactions of downstream tasks. Our benchmark results on three property types of real-world transactions show that DoRA significantly outperforms the SSL baselines for tabular data, the graph-based methods, and the supervised approaches in the few-shot scenarios by at least 7.6% for MAPE, 11.59% for MAE, and 3.34% for HR10%. We expect DoRA to be useful to other financial practitioners with similar marketplace applications who need general models for properties that are newly built and have limited records. The source code is available at https://github.com/wwweiwei/DoRA.

Wei-Yao Wang, Wen-Chih Peng

In recent years, social media has enabled users to get exposed to a myriad of misinformation and disinformation; thus, misinformation has attracted a great deal of attention in research fields and as a social issue. To address the problem, we propose a framework, Pre-CoFact, composed of two pre-trained models for extracting features from text and images, and multiple co-attention networks for fusing the same modality but different sources and different modalities. Besides, we adopt the ensemble method by using different pre-trained models in Pre-CoFact to achieve better performance. We further illustrate the effectiveness from the ablation study and examine different pre-trained models for comparison. Our team, Yao, won the fifth prize (F1-score: 74.585\%) in the Factify challenge hosted by De-Factify @ AAAI 2022, which demonstrates that our model achieved competitive performance without using auxiliary tasks or extra information. The source code of our work is publicly available at https://github.com/wywyWang/Multi-Modal-Fact-Verification-2021

Wei-Yao Wang, Teng-Fong Chan, Hui-Kuo Yang et al.

Identifying significant shots in a rally is important for evaluating players' performance in badminton matches. While there are several studies that have quantified player performance in other sports, analyzing badminton data is remained untouched. In this paper, we introduce a badminton language to fully describe the process of the shot and propose a deep learning model composed of a novel short-term extractor and a long-term encoder for capturing a shot-by-shot sequence in a badminton rally by framing the problem as predicting a rally result. Our model incorporates an attention mechanism to enable the transparency of the action sequence to the rally result, which is essential for badminton experts to gain interpretable predictions. Experimental evaluation based on a real-world dataset demonstrates that our proposed model outperforms the strong baselines. The source code is publicly available at https://github.com/yao0510/Shot-Influence.

Wei-Yao Wang, Wei-Wei Du, Derek Xu et al.

Self-supervised learning (SSL) has been incorporated into many state-of-the-art models in various domains, where SSL defines pretext tasks based on unlabeled datasets to learn contextualized and robust representations. Recently, SSL has become a new trend in exploring the representation learning capability in the realm of tabular data, which is more challenging due to not having explicit relations for learning descriptive representations. This survey aims to systematically review and summarize the recent progress and challenges of SSL for non-sequential tabular data (SSL4NS-TD). We first present a formal definition of NS-TD and clarify its correlation to related studies. Then, these approaches are categorized into three groups - predictive learning, contrastive learning, and hybrid learning, with their motivations and strengths of representative methods in each direction. Moreover, application issues of SSL4NS-TD are presented, including automatic data engineering, cross-table transferability, and domain knowledge integration. In addition, we elaborate on existing benchmarks and datasets for NS-TD applications to analyze the performance of existing tabular models. Finally, we discuss the challenges of SSL4NS-TD and provide potential directions for future research. We expect our work to be useful in terms of encouraging more research on lowering the barrier to entry SSL for the tabular domain, and of improving the foundations for implicit tabular data.

Wenyan Cong, Hanqing Zhu, Kevin Wang et al.

Efficiently reconstructing 3D scenes from monocular video remains a core challenge in computer vision, vital for applications in virtual reality, robotics, and scene understanding. Recently, frame-by-frame progressive reconstruction without camera poses is commonly adopted, incurring high computational overhead and compounding errors when scaling to longer videos. To overcome these issues, we introduce VideoLifter, a novel video-to-3D pipeline that leverages a local-to-global strategy on a fragment basis, achieving both extreme efficiency and SOTA quality. Locally, VideoLifter leverages learnable 3D priors to register fragments, extracting essential information for subsequent 3D Gaussian initialization with enforced inter-fragment consistency and optimized efficiency. Globally, it employs a tree-based hierarchical merging method with key frame guidance for inter-fragment alignment, pairwise merging with Gaussian point pruning, and subsequent joint optimization to ensure global consistency while efficiently mitigating cumulative errors. This approach significantly accelerates the reconstruction process, reducing training time by over 82% while holding better visual quality than current SOTA methods.

Wei-Yao Wang, Yu-Chieh Chang, Wen-Chih Peng

With the improvements in generative models, the issues of producing hallucinations in various domains (e.g., law, writing) have been brought to people's attention due to concerns about misinformation. In this paper, we focus on neural fake news, which refers to content generated by neural networks aiming to mimic the style of real news to deceive people. To prevent harmful disinformation spreading fallaciously from malicious social media (e.g., content farms), we propose a novel verification framework, Style-News, using publisher metadata to imply a publisher's template with the corresponding text types, political stance, and credibility. Based on threat modeling aspects, a style-aware neural news generator is introduced as an adversary for generating news content conditioning for a specific publisher, and style and source discriminators are trained to defend against this attack by identifying which publisher the style corresponds with, and discriminating whether the source of the given news is human-written or machine-generated. To evaluate the quality of the generated content, we integrate various dimensional metrics (language fluency, content preservation, and style adherence) and demonstrate that Style-News significantly outperforms the previous approaches by a margin of 0.35 for fluency, 15.24 for content, and 0.38 for style at most. Moreover, our discriminative model outperforms state-of-the-art baselines in terms of publisher prediction (up to 4.64%) and neural fake news detection (+6.94% $\sim$ 31.72%).

Ching Chang, Chiao-Tung Chan, Wei-Yao Wang et al.

Multivariate time-series data in fields like healthcare and industry are informative but challenging due to high dimensionality and lack of labels. Recent self-supervised learning methods excel in learning rich representations without labels but struggle with disentangled embeddings and inductive bias issues like transformation-invariance. To address these challenges, we introduce TimeDRL, a framework for multivariate time-series representation learning with dual-level disentangled embeddings. TimeDRL features: (i) disentangled timestamp-level and instance-level embeddings using a [CLS] token strategy; (ii) timestamp-predictive and instance-contrastive tasks for representation learning; and (iii) avoidance of augmentation methods to eliminate inductive biases. Experiments on forecasting and classification datasets show TimeDRL outperforms existing methods, with further validation in semi-supervised settings with limited labeled data.

Weiyao Wang, Pierre Gleize, Hao Tang et al.

Neural Radiance Fields (NeRF) exhibit remarkable performance for Novel View Synthesis (NVS) given a set of 2D images. However, NeRF training requires accurate camera pose for each input view, typically obtained by Structure-from-Motion (SfM) pipelines. Recent works have attempted to relax this constraint, but they still often rely on decent initial poses which they can refine. Here we aim at removing the requirement for pose initialization. We present Incremental CONfidence (ICON), an optimization procedure for training NeRFs from 2D video frames. ICON only assumes smooth camera motion to estimate initial guess for poses. Further, ICON introduces ``confidence": an adaptive measure of model quality used to dynamically reweight gradients. ICON relies on high-confidence poses to learn NeRF, and high-confidence 3D structure (as encoded by NeRF) to learn poses. We show that ICON, without prior pose initialization, achieves superior performance in both CO3D and HO3D versus methods which use SfM pose.

Qiyu Wu, Shuyang Cui, Satoshi Hayakawa et al.

Multimodal retrieval, which seeks to retrieve relevant content across modalities such as text or image, supports applications from AI search to contents production. Despite the success of separate-encoder approaches like CLIP align modality-specific embeddings with contrastive learning, recent multimodal large language models (MLLMs) enable a unified encoder that directly processes composed inputs. While flexible and advanced, we identify that unified encoders trained with conventional contrastive learning are prone to learn modality shortcut, leading to poor robustness under distribution shifts. We propose a modality composition awareness framework to mitigate this issue. Concretely, a preference loss enforces multimodal embeddings to outperform their unimodal counterparts, while a composition regularization objective aligns multimodal embeddings with prototypes composed from its unimodal parts. These objectives explicitly model structural relationships between the composed representation and its unimodal counterparts. Experiments on various benchmarks show gains in out-of-distribution retrieval, highlighting modality composition awareness as a effective principle for robust composed multimodal retrieval when utilizing MLLMs as the unified encoder.

Kuang-Da Wang, Zhao Wang, Yotaro Shimose et al.

Witnessed by the recent advancements on leveraging LLM for coding and multimodal understanding, we present WebGen-V, a new benchmark and framework for instruction-to-HTML generation that enhances both data quality and evaluation granularity. WebGen-V contributes three key innovations: (1) an unbounded and extensible agentic crawling framework that continuously collects real-world webpages and can leveraged to augment existing benchmarks; (2) a structured, section-wise data representation that integrates metadata, localized UI screenshots, and JSON-formatted text and image assets, explicit alignment between content, layout, and visual components for detailed multimodal supervision; and (3) a section-level multimodal evaluation protocol aligning text, layout, and visuals for high-granularity assessment. Experiments with state-of-the-art LLMs and ablation studies validate the effectiveness of our structured data and section-wise evaluation, as well as the contribution of each component. To the best of our knowledge, WebGen-V is the first work to enable high-granularity agentic crawling and evaluation for instruction-to-HTML generation, providing a unified pipeline from real-world data acquisition and webpage generation to structured multimodal assessment.

Wei-Yao Wang, Kazuya Tateishi, Qiyu Wu et al.

Multimodal representation learning models have demonstrated successful operation across complex tasks, and the integration of vision-language models (VLMs) has further enabled embedding models with instruction-following capabilities. However, existing embedding models lack visual-interactive capabilities to specify regions of interest from users (e.g., point, bounding box, mask), which have been explored in generative models to broaden their human-interactive applicability. Equipping embedding models with visual interactions not only would unlock new applications with localized grounding of user intent, which remains unexplored, but also enable the models to learn entity-level information within images to complement their global representations for conventional embedding tasks. In this paper, we propose a novel Visual-InteRactive Text-Image Universal Embedder (VIRTUE) that extends the capabilities of the segmentation model and the vision-language model to the realm of representation learning. In VIRTUE, the segmentation model can process visual prompts that pinpoint specific regions within an image, thereby enabling the embedder to handle complex and ambiguous scenarios more precisely. To evaluate the visual-interaction ability of VIRTUE, we introduce a large-scale Segmentation-and-Scene Caption Retrieval (SCaR) benchmark comprising 1M samples that aims to retrieve the text caption by jointly considering the entity with a specific object and image scene. VIRTUE consistently achieves a state-of-the-art performance with significant improvements across 36 universal MMEB (3.1%-8.5%) and five visual-interactive SCaR (15.2%-20.3%) tasks.

Yen-Che Chien, Kuang-Da Wang, Wei-Yao Wang et al.

Recent advances in autonomous digital agents from industry (e.g., Manus AI and Gemini's research mode) highlight potential for structured tasks by autonomous decision-making and task decomposition; however, it remains unclear to what extent the agent-based systems can improve multimodal web data productivity. We study this in the realm of journalism, which requires iterative planning, interpretation, and contextual reasoning from multimodal raw contents to form a well structured news. We introduce NEWSAGENT, a benchmark for evaluating how agents can automatically search available raw contents, select desired information, and edit and rephrase to form a news article by accessing core journalistic functions. Given a writing instruction and firsthand data as how a journalist initiates a news draft, agents are tasked to identify narrative perspectives, issue keyword-based queries, retrieve historical background, and generate complete articles. Unlike typical summarization or retrieval tasks, essential context is not directly available and must be actively discovered, reflecting the information gaps faced in real-world news writing. NEWSAGENT includes 6k human-verified examples derived from real news, with multimodal contents converted to text for broad model compatibility. We evaluate open- and closed-sourced LLMs with commonly-used agentic frameworks on NEWSAGENT, which shows that agents are capable of retrieving relevant facts but struggling with planning and narrative integration. We believe that NEWSAGENT serves a realistic testbed for iterating and evaluating agent capabilities in terms of multimodal web data manipulation to real-world productivity.

Yung-Chien Wang, Kuang-Da Wang, Wei-Yao Wang et al.

Tabular data serve as a fundamental and ubiquitous representation of structured information in numerous real-world applications, e.g., finance and urban planning. In the realm of tabular imbalanced applications, data imbalance has been investigated in classification tasks with insufficient instances in certain labels, causing the model's ineffective generalizability. However, the imbalance issue of tabular regression tasks is underexplored, and yet is critical due to unclear boundaries for continuous labels and simplifying assumptions in existing imbalance regression work, which often rely on known and balanced test distributions. Such assumptions may not hold in practice and can lead to performance degradation. To address these issues, we propose MATI: Mixture Experts with Test-Time Self-Supervised Aggregation for Tabular Imbalance Regression, featuring two key innovations: (i) the Region-Aware Mixture Expert, which adopts a Gaussian Mixture Model to capture the underlying related regions. The statistical information of each Gaussian component is then used to synthesize and train region-specific experts to capture the unique characteristics of their respective regions. (ii) Test-Time Self-Supervised Expert Aggregation, which dynamically adjusts region expert weights based on test data features to reinforce expert adaptation across varying test distributions. We evaluated MATI on four real-world tabular imbalance regression datasets, including house pricing, bike sharing, and age prediction. To reflect realistic deployment scenarios, we adopted three types of test distributions: a balanced distribution with uniform target frequencies, a normal distribution that follows the training data, and an inverse distribution that emphasizes rare target regions. On average across these three test distributions, MATI achieved a 7.1% improvement in MAE compared to existing methods.

Kuang-Da Wang, Wei-Yao Wang, Ping-Chun Hsieh et al.

In the dynamic and rapid tactic involvements of turn-based sports, badminton stands out as an intrinsic paradigm that requires alter-dependent decision-making of players. While the advancement of learning from offline expert data in sequential decision-making has been witnessed in various domains, how to rally-wise imitate the behaviors of human players from offline badminton matches has remained underexplored. Replicating opponents' behavior benefits players by allowing them to undergo strategic development with direction before matches. However, directly applying existing methods suffers from the inherent hierarchy of the match and the compounding effect due to the turn-based nature of players alternatively taking actions. In this paper, we propose RallyNet, a novel hierarchical offline imitation learning model for badminton player behaviors: (i) RallyNet captures players' decision dependencies by modeling decision-making processes as a contextual Markov decision process. (ii) RallyNet leverages the experience to generate context as the agent's intent in the rally. (iii) To generate more realistic behavior, RallyNet leverages Geometric Brownian Motion (GBM) to model the interactions between players by introducing a valuable inductive bias for learning player behaviors. In this manner, RallyNet links player intents with interaction models with GBM, providing an understanding of interactions for sports analytics. We extensively validate RallyNet with the largest available real-world badminton dataset consisting of men's and women's singles, demonstrating its ability to imitate player behaviors. Results reveal RallyNet's superiority over offline imitation learning methods and state-of-the-art turn-based approaches, outperforming them by at least 16% in mean rule-based agent normalization score. Furthermore, we discuss various practical use cases to highlight RallyNet's applicability.

Wei-Yao Wang, Wen-Chih Peng, Wei Wang et al.

Agent forecasting systems have been explored to investigate agent patterns and improve decision-making in various domains, e.g., pedestrian predictions and marketing bidding. Badminton represents a fascinating example of a multifaceted turn-based sport, requiring both sophisticated tactic developments and alternate-dependent decision-making. Recent deep learning approaches for player tactic forecasting in badminton show promising performance partially attributed to effective reasoning about rally-player interactions. However, a critical obstacle lies in the unclear functionality of which features are learned for simulating players' behaviors by black-box models, where existing explainers are not equipped with turn-based and multi-output attributions. To bridge this gap, we propose a turn-based feature attribution approach, ShuttleSHAP, for analyzing forecasting models in badminton based on variants of Shapley values. ShuttleSHAP is a model-agnostic explainer that aims to quantify contribution by not only temporal aspects but also player aspects in terms of multifaceted cues. Incorporating the proposed analysis tool into the state-of-the-art turn-based forecasting model on the benchmark dataset reveals that it is, in fact, insignificant to reason about past strokes, while conventional sequential models have greater impacts. Instead, players' styles influence the models for the future simulation of a rally. On top of that, we investigate and discuss the causal analysis of these findings and demonstrate the practicability with local analysis.

Wei-Yao Wang, Hong-Han Shuai, Kai-Shiang Chang et al.

The increasing demand for analyzing the insights in sports has stimulated a line of productive studies from a variety of perspectives, e.g., health state monitoring, outcome prediction. In this paper, we focus on objectively judging what and where to return strokes, which is still unexplored in turn-based sports. By formulating stroke forecasting as a sequence prediction task, existing works can tackle the problem but fail to model information based on the characteristics of badminton. To address these limitations, we propose a novel Position-aware Fusion of Rally Progress and Player Styles framework (ShuttleNet) that incorporates rally progress and information of the players by two modified encoder-decoder extractors. Moreover, we design a fusion network to integrate rally contexts and contexts of the players by conditioning on information dependency and different positions. Extensive experiments on the badminton dataset demonstrate that ShuttleNet significantly outperforms the state-of-the-art methods and also empirically validates the feasibility of each component in ShuttleNet. On top of that, we provide an analysis scenario for the stroke forecasting problem.

Weiyao Wang, Marin Kobilarov, Gregory D. Hager

Recent advances in latent space dynamics model from pixels show promising progress in vision-based model predictive control (MPC). However, executing MPC in real time can be challenging due to its intensive computational cost in each timestep. We propose to introduce additional learning objectives to enforce that the learned latent space is proportional derivative controllable. In execution time, the simple PD-controller can be applied directly to the latent space encoded from pixels, to produce simple and effective control to systems with visual observations. We show that our method outperforms baseline methods to produce robust goal reaching and trajectory tracking in various environments.

Weiyao Wang, Matt Feiszli, Heng Wang et al.

Current state-of-the-art object detection and segmentation methods work well under the closed-world assumption. This closed-world setting assumes that the list of object categories is available during training and deployment. However, many real-world applications require detecting or segmenting novel objects, i.e., object categories never seen during training. In this paper, we present, UVO (Unidentified Video Objects), a new benchmark for open-world class-agnostic object segmentation in videos. Besides shifting the problem focus to the open-world setup, UVO is significantly larger, providing approximately 8 times more videos compared with DAVIS, and 7 times more mask (instance) annotations per video compared with YouTube-VOS and YouTube-VIS. UVO is also more challenging as it includes many videos with crowded scenes and complex background motions. We demonstrated that UVO can be used for other applications, such as object tracking and super-voxel segmentation, besides open-world object segmentation. We believe that UVo is a versatile testbed for researchers to develop novel approaches for open-world class-agnostic object segmentation, and inspires new research directions towards a more comprehensive video understanding beyond classification and detection.

Mike Zheng Shou, Stan Weixian Lei, Weiyao Wang et al.

This paper presents a novel task together with a new benchmark for detecting generic, taxonomy-free event boundaries that segment a whole video into chunks. Conventional work in temporal video segmentation and action detection focuses on localizing pre-defined action categories and thus does not scale to generic videos. Cognitive Science has known since last century that humans consistently segment videos into meaningful temporal chunks. This segmentation happens naturally, without pre-defined event categories and without being explicitly asked to do so. Here, we repeat these cognitive experiments on mainstream CV datasets; with our novel annotation guideline which addresses the complexities of taxonomy-free event boundary annotation, we introduce the task of Generic Event Boundary Detection (GEBD) and the new benchmark Kinetics-GEBD. Our Kinetics-GEBD has the largest number of boundaries (e.g. 32 of ActivityNet, 8 of EPIC-Kitchens-100) which are in-the-wild, taxonomy-free, cover generic event change, and respect human perception diversity. We view GEBD as an important stepping stone towards understanding the video as a whole, and believe it has been previously neglected due to a lack of proper task definition and annotations. Through experiment and human study we demonstrate the value of the annotations. Further, we benchmark supervised and un-supervised GEBD approaches on the TAPOS dataset and our Kinetics-GEBD. We release our annotations and baseline codes at CVPR'21 LOVEU Challenge: https://sites.google.com/view/loveucvpr21.

Qihao Liu, Weichao Qiu, Weiyao Wang et al.

We propose an unsupervised vision-based system to estimate the joint configurations of the robot arm from a sequence of RGB or RGB-D images without knowing the model a priori, and then adapt it to the task of category-independent articulated object pose estimation. We combine a classical geometric formulation with deep learning and extend the use of epipolar constraint to multi-rigid-body systems to solve this task. Given a video sequence, the optical flow is estimated to get the pixel-wise dense correspondences. After that, the 6D pose is computed by a modified PnP algorithm. The key idea is to leverage the geometric constraints and the constraint between multiple frames. Furthermore, we build a synthetic dataset with different kinds of robots and multi-joint articulated objects for the research of vision-based robot control and robotic vision. We demonstrate the effectiveness of our method on three benchmark datasets and show that our method achieves higher accuracy than the state-of-the-art supervised methods in estimating joint angles of robot arms and articulated objects.

Wei-Yao Wang, Kai-Shiang Chang, Yu-Chien Tang

This paper provides a method to classify sentiment with robust model based ensemble methods. We preprocess tweet data to enhance coverage of tokenizer. To reduce domain bias, we first train tweet dataset for pre-trained language model. Besides, each classifier has its strengths and weakness, we leverage different types of models with ensemble methods: average and power weighted sum. From the experiments, we show that our approach has achieved positive effect for sentiment classification. Our system reached third place among 26 teams from the evaluation in SocialNLP 2020 EmotionGIF competition.

Weiyao Wang, Du Tran, Matt Feiszli

Consider end-to-end training of a multi-modal vs. a single-modal network on a task with multiple input modalities: the multi-modal network receives more information, so it should match or outperform its single-modal counterpart. In our experiments, however, we observe the opposite: the best single-modal network always outperforms the multi-modal network. This observation is consistent across different combinations of modalities and on different tasks and benchmarks. This paper identifies two main causes for this performance drop: first, multi-modal networks are often prone to overfitting due to increased capacity. Second, different modalities overfit and generalize at different rates, so training them jointly with a single optimization strategy is sub-optimal. We address these two problems with a technique we call Gradient Blending, which computes an optimal blend of modalities based on their overfitting behavior. We demonstrate that Gradient Blending outperforms widely-used baselines for avoiding overfitting and achieves state-of-the-art accuracy on various tasks including human action recognition, ego-centric action recognition, and acoustic event detection.