Haiyang Wang

Haiyang Wang, Chen Shi, Shaoshuai Shi et al. · pku

Designing an efficient yet deployment-friendly 3D backbone to handle sparse point clouds is a fundamental problem in 3D perception. Compared with the customized sparse convolution, the attention mechanism in Transformers is more appropriate for flexibly modeling long-range relationships and is easier to be deployed in real-world applications. However, due to the sparse characteristics of point clouds, it is non-trivial to apply a standard transformer on sparse points. In this paper, we present Dynamic Sparse Voxel Transformer (DSVT), a single-stride window-based voxel Transformer backbone for outdoor 3D perception. In order to efficiently process sparse points in parallel, we propose Dynamic Sparse Window Attention, which partitions a series of local regions in each window according to its sparsity and then computes the features of all regions in a fully parallel manner. To allow the cross-set connection, we design a rotated set partitioning strategy that alternates between two partitioning configurations in consecutive self-attention layers. To support effective downsampling and better encode geometric information, we also propose an attention-style 3D pooling module on sparse points, which is powerful and deployment-friendly without utilizing any customized CUDA operations. Our model achieves state-of-the-art performance with a broad range of 3D perception tasks. More importantly, DSVT can be easily deployed by TensorRT with real-time inference speed (27Hz). Code will be available at \url{https://github.com/Haiyang-W/DSVT}.

Haiyang Wang, Hao Tang, Shaoshuai Shi et al. · pku

Jointly processing information from multiple sensors is crucial to achieving accurate and robust perception for reliable autonomous driving systems. However, current 3D perception research follows a modality-specific paradigm, leading to additional computation overheads and inefficient collaboration between different sensor data. In this paper, we present an efficient multi-modal backbone for outdoor 3D perception named UniTR, which processes a variety of modalities with unified modeling and shared parameters. Unlike previous works, UniTR introduces a modality-agnostic transformer encoder to handle these view-discrepant sensor data for parallel modal-wise representation learning and automatic cross-modal interaction without additional fusion steps. More importantly, to make full use of these complementary sensor types, we present a novel multi-modal integration strategy by both considering semantic-abundant 2D perspective and geometry-aware 3D sparse neighborhood relations. UniTR is also a fundamentally task-agnostic backbone that naturally supports different 3D perception tasks. It sets a new state-of-the-art performance on the nuScenes benchmark, achieving +1.1 NDS higher for 3D object detection and +12.0 higher mIoU for BEV map segmentation with lower inference latency. Code will be available at https://github.com/Haiyang-W/UniTR .

Haiyang Wang, Shaoshuai Shi, Ze Yang et al. · pku

As a fundamental problem in computer vision, 3D object detection is experiencing rapid growth. To extract the point-wise features from the irregularly and sparsely distributed points, previous methods usually take a feature grouping module to aggregate the point features to an object candidate. However, these methods have not yet leveraged the surface geometry of foreground objects to enhance grouping and 3D box generation. In this paper, we propose the RBGNet framework, a voting-based 3D detector for accurate 3D object detection from point clouds. In order to learn better representations of object shape to enhance cluster features for predicting 3D boxes, we propose a ray-based feature grouping module, which aggregates the point-wise features on object surfaces using a group of determined rays uniformly emitted from cluster centers. Considering the fact that foreground points are more meaningful for box estimation, we design a novel foreground biased sampling strategy in downsample process to sample more points on object surfaces and further boost the detection performance. Our model achieves state-of-the-art 3D detection performance on ScanNet V2 and SUN RGB-D with remarkable performance gains. Code will be available at https://github.com/Haiyang-W/RBGNet.

Haiyang Wang, Lihe Ding, Shaocong Dong et al. · pku

We present a novel two-stage fully sparse convolutional 3D object detection framework, named CAGroup3D. Our proposed method first generates some high-quality 3D proposals by leveraging the class-aware local group strategy on the object surface voxels with the same semantic predictions, which considers semantic consistency and diverse locality abandoned in previous bottom-up approaches. Then, to recover the features of missed voxels due to incorrect voxel-wise segmentation, we build a fully sparse convolutional RoI pooling module to directly aggregate fine-grained spatial information from backbone for further proposal refinement. It is memory-and-computation efficient and can better encode the geometry-specific features of each 3D proposal. Our model achieves state-of-the-art 3D detection performance with remarkable gains of +\textit{3.6\%} on ScanNet V2 and +\textit{2.6}\% on SUN RGB-D in term of mAP@0.25. Code will be available at https://github.com/Haiyang-W/CAGroup3D.

Hao Yang, Haiyang Wang, Di Dai et al. · pku

Pre-training is crucial in 3D-related fields such as autonomous driving where point cloud annotation is costly and challenging. Many recent studies on point cloud pre-training, however, have overlooked the issue of incompleteness, where only a fraction of the points are captured by LiDAR, leading to ambiguity during the training phase. On the other hand, images offer more comprehensive information and richer semantics that can bolster point cloud encoders in addressing the incompleteness issue inherent in point clouds. Yet, incorporating images into point cloud pre-training presents its own challenges due to occlusions, potentially causing misalignments between points and pixels. In this work, we propose PRED, a novel image-assisted pre-training framework for outdoor point clouds in an occlusion-aware manner. The main ingredient of our framework is a Birds-Eye-View (BEV) feature map conditioned semantic rendering, leveraging the semantics of images for supervision through neural rendering. We further enhance our model's performance by incorporating point-wise masking with a high mask ratio (95%). Extensive experiments demonstrate PRED's superiority over prior point cloud pre-training methods, providing significant improvements on various large-scale datasets for 3D perception tasks. Codes will be available at https://github.com/PRED4pc/PRED.

Jiandong Jin, Xiao Wang, Qian Zhu et al.

Pedestrian Attribute Recognition (PAR) is one of the indispensable tasks in human-centered research. However, existing datasets neglect different domains (e.g., environments, times, populations, and data sources), only conducting simple random splits, and the performance of these datasets has already approached saturation. In the past five years, no large-scale dataset has been opened to the public. To address this issue, this paper proposes a new large-scale, cross-domain pedestrian attribute recognition dataset to fill the data gap, termed MSP60K. It consists of 60,122 images and 57 attribute annotations across eight scenarios. Synthetic degradation is also conducted to further narrow the gap between the dataset and real-world challenging scenarios. To establish a more rigorous benchmark, we evaluate 17 representative PAR models under both random and cross-domain split protocols on our dataset. Additionally, we propose an innovative Large Language Model (LLM) augmented PAR framework, named LLM-PAR. This framework processes pedestrian images through a Vision Transformer (ViT) backbone to extract features and introduces a multi-embedding query Transformer to learn partial-aware features for attribute classification. Significantly, we enhance this framework with LLM for ensemble learning and visual feature augmentation. Comprehensive experiments across multiple PAR benchmark datasets have thoroughly validated the efficacy of our proposed framework. The dataset and source code accompanying this paper will be made publicly available at \url{https://github.com/Event-AHU/OpenPAR}.

Lin Mu, Haiyang Wang, Li Ni et al.

Low-Rank Adaptation (LoRA) enables parameter-efficient fine-tuning of Large Language Models (LLMs), and recent Mixture-of-Experts (MoE) extensions further enhance flexibility by dynamically combining multiple LoRA experts. However, existing MoE-augmented LoRA methods assume that experts operate independently, often leading to unstable routing, expert dominance. In this paper, we propose \textbf{TalkLoRA}, a communication-aware MoELoRA framework that relaxes this independence assumption by introducing expert-level communication prior to routing. TalkLoRA equips low-rank experts with a lightweight Talking Module that enables controlled information exchange across expert subspaces, producing a more robust global signal for routing. Theoretically, we show that expert communication smooths routing dynamics by mitigating perturbation amplification while strictly generalizing existing MoELoRA architectures. Empirically, TalkLoRA consistently outperforms vanilla LoRA and MoELoRA across diverse language understanding and generation tasks, achieving higher parameter efficiency and more balanced expert routing under comparable parameter budgets. These results highlight structured expert communication as a principled and effective enhancement for MoE-based parameter-efficient adaptation. Code is available at https://github.com/why0129/TalkLoRA.

Qixing Zhou, Jiacheng Zhang, Haiyang Wang et al.

Agents powered by large language models (LLMs) are increasingly adopted in the software industry, contributing code as collaborators or even autonomous developers. As their presence grows, it becomes important to assess the current boundaries of their coding abilities. Existing agentic coding benchmarks, however, cover a limited task scope, e.g., bug fixing within a single pull request (PR), and often rely on non-executable evaluations or lack an automated approach for continually updating the evaluation coverage. To address such issues, we propose FeatureBench, a benchmark designed to evaluate agentic coding performance in end-to-end, feature-oriented software development. FeatureBench incorporates an execution-based evaluation protocol and a scalable test-driven method that automatically derives tasks from code repositories with minimal human effort. By tracing from unit tests along a dependency graph, our approach can identify feature-level coding tasks spanning multiple commits and PRs scattered across the development timeline, while ensuring the proper functioning of other features after the separation. Using this framework, we curated 200 challenging evaluation tasks and 3825 executable environments from 24 open-source repositories in the first version of our benchmark. Empirical evaluation reveals that the state-of-the-art agentic model, such as Claude 4.5 Opus, which achieves a 74.4% resolved rate on SWE-bench, succeeds on only 11.0% of tasks, opening new opportunities for advancing agentic coding. Moreover, benefiting from our automated task collection toolkit, FeatureBench can be easily scaled and updated over time to mitigate data leakage. The inherent verifiability of constructed environments also makes our method potentially valuable for agent training.

Dingbang Wu, Rui Hao, Haiyang Wang et al.

We present MobileGym, a browser-hosted, lightweight, fully controllable environment for everyday mobile use, targeting interaction fidelity without replicating proprietary backends. It enables two capabilities previously out of reach for everyday apps: verifiable outcome signals through deterministic state-based judging over structured JSON state, and scalable online RL through low-cost parallel rollouts. The full environment state is captured, configured, forked, and compared as structured JSON, and a single server can host hundreds of parallel instances, with about 400 MB memory per instance and about 3 s cold start. A layered state model and a declarative task-definition framework keep state programmability and task creation practical at scale, and a single programmatic judging mechanism delivers both deterministic evaluation verdicts and dense RL rewards. The accompanying MobileGym-Bench provides 416 parameterized task templates, including 256 test and 160 train templates, over 28 apps, with deterministic judges and a structured AnswerSheet protocol that avoids free-text matching failures. In a Sim-to-Real case study, GRPO on Qwen3-VL-4B-Instruct gains +12.8 percentage points on the 256-task test set, and on a 59-task real-device signal subset, real-device execution retains 95.1% of the simulation-side training gain. Project page: https://mobilegym.github.io.

Yusong Lin, Xinyuan Liang, Haiyang Wang et al.

Large language model agents are increasingly envisioned as always-on personal assistants with access to anything relevant in the user's digital world. Yet current systems operate over only narrow slices of that world, limiting context-sensitive reasoning and effective assistance. Existing benchmarks similarly provide only partial user state and therefore fail to capture performance in such a broad, always-on setting. To address this gap, we introduce Claw-Anything, a benchmark that expands agent context along three dimensions: long-horizon activity histories, interdependent backend services, and integrated GUI and CLI interaction across multiple devices. To instantiate this setting, we simulate months of user activity through multi-round event injection, producing complex world states and realistic noise, including irrelevant events and conflicting signals. Agents must reason over rich contextual environments while remaining robust to such noise. This expanded scope also enables the evaluation of proactive assistance, requiring agents to anticipate user needs and deliver timely recommendations. Experiments show that GPT-5.5 achieves only 34.5% pass@1, substantially below prior benchmarks, underscoring a gap between current agent capabilities and the demands of always-on personal assistance. Alongside the benchmark, we release an automated data-generation pipeline that yields 2,000 training environments and improves the base model by 23.7%, demonstrating its utility of scalable data infrastructure.

Feng Xie, Zhong Zhang, Xuechen Zhao et al.

The ongoing COVID-19 pandemic has caused immeasurable losses for people worldwide. To contain the spread of the virus and further alleviate the crisis, various health policies (e.g., stay-at-home orders) have been issued which spark heated discussions as users turn to share their attitudes on social media. In this paper, we consider a more realistic scenario on stance detection (i.e., cross-target and zero-shot settings) for the pandemic and propose an adversarial learning-based stance classifier to automatically identify the public's attitudes toward COVID-19-related health policies. Specifically, we adopt adversarial learning that allows the model to train on a large amount of labeled data and capture transferable knowledge from source topics, so as to enable generalize to the emerging health policies with sparse labeled data. To further enhance the model's deeper understanding, we incorporate policy descriptions as external knowledge into the model. Meanwhile, a GeoEncoder is designed which encourages the model to capture unobserved background factors specified by each region and then represent them as non-text information. We evaluate the performance of a broad range of baselines on the stance detection task for COVID-19-related health policies, and experimental results show that our proposed method achieves state-of-the-art performance in both cross-target and zero-shot settings.

Haiyang Wang, Luca Barletta, Alex Dytso

We study the amplitude-constrained additive white Gaussian noise channel. It is well known that the capacity-achieving input distribution for this channel is discrete and supported on finitely many points. The best known bounds show that the support size of the capacity-achieving distribution is lower-bounded by a term of order $A$ and upper-bounded by a term of order $A^2$, where $A$ denotes the amplitude constraint. It was conjectured in [1] that the linear scaling is optimal. In this work, we establish a new lower bound of order $A\sqrt{\log A}$, improving the known bound and ruling out the conjectured linear scaling. To obtain this result, we quantify the fact that the capacity-achieving output distribution is close to the uniform distribution in the interior of the amplitude constraint. Next, we introduce a wrapping operation that maps the problem to a compact domain and develop a theory of best approximation of the uniform distribution by finite Gaussian mixtures. These approximation bounds are then combined with stability properties of capacity-achieving distributions to yield the final support-size lower bound.

Hao Tang, Chenwei Xie, Haiyang Wang et al. · pku

Generalist models have achieved remarkable success in both language and vision-language tasks, showcasing the potential of unified modeling. However, effectively integrating fine-grained perception tasks like detection and segmentation into these models remains a significant challenge. This is primarily because these tasks often rely heavily on task-specific designs and architectures that can complicate the modeling process. To address this challenge, we present \ours, a framework that \textbf{U}nifies \textbf{F}ine-grained visual perception tasks through an \textbf{O}pen-ended language interface. By transforming all perception targets into the language space, \ours unifies object-level detection, pixel-level segmentation, and image-level vision-language tasks into a single model. Additionally, we introduce a novel embedding retrieval approach that relies solely on the language interface to support segmentation tasks. Our framework bridges the gap between fine-grained perception and vision-language tasks, significantly simplifying architectural design and training strategies while achieving comparable or superior performance to methods with intricate task-specific designs. After multi-task training on five standard visual perception datasets, \ours outperforms the previous state-of-the-art generalist models by 12.3 mAP on COCO instance segmentation and 3.3 mIoU on ADE20K semantic segmentation. Furthermore, our method seamlessly integrates with existing MLLMs, effectively combining fine-grained perception capabilities with their advanced language abilities, thereby enabling more challenging tasks such as reasoning segmentation. Code and models are available at https://github.com/nnnth/UFO.

Haiyang Wang, Yue Fan, Muhammad Ferjad Naeem et al. · pku

Transformers have become the predominant architecture in foundation models due to their excellent performance across various domains. However, the substantial cost of scaling these models remains a significant concern. This problem arises primarily from their dependence on a fixed number of parameters within linear projections. When architectural modifications (e.g., channel dimensions) are introduced, the entire model typically requires retraining from scratch. As model sizes continue growing, this strategy results in increasingly high computational costs and becomes unsustainable. To overcome this problem, we introduce TokenFormer, a natively scalable architecture that leverages the attention mechanism not only for computations among input tokens but also for interactions between tokens and model parameters, thereby enhancing architectural flexibility. By treating model parameters as tokens, we replace all the linear projections in Transformers with our token-parameter attention layer, where input tokens act as queries and model parameters as keys and values. This reformulation allows for progressive and efficient scaling without necessitating retraining from scratch. Our model scales from 124M to 1.4B parameters by incrementally adding new key-value parameter pairs, achieving performance comparable to Transformers trained from scratch while greatly reducing training costs. Code and models are available at https://github.com/Haiyang-W/TokenFormer.

Xiao Wang, Haiyang Wang, Shiao Wang et al.

Existing pedestrian attribute recognition methods are generally developed based on RGB frame cameras. However, these approaches are constrained by the limitations of RGB cameras, such as sensitivity to lighting conditions and motion blur, which hinder their performance. Furthermore, current attribute recognition primarily focuses on analyzing pedestrians' external appearance and clothing, lacking an exploration of emotional dimensions. In this paper, we revisit these issues and propose a novel multi-modal RGB-Event attribute recognition task by drawing inspiration from the advantages of event cameras in low-light, high-speed, and low-power consumption. Specifically, we introduce the first large-scale multi-modal pedestrian attribute recognition dataset, termed EventPAR, comprising 100K paired RGB-Event samples that cover 50 attributes related to both appearance and six human emotions, diverse scenes, and various seasons. By retraining and evaluating mainstream PAR models on this dataset, we establish a comprehensive benchmark and provide a solid foundation for future research in terms of data and algorithmic baselines. In addition, we propose a novel RWKV-based multi-modal pedestrian attribute recognition framework, featuring an RWKV visual encoder and an asymmetric RWKV fusion module. Extensive experiments are conducted on our proposed dataset as well as two simulated datasets (MARS-Attribute and DukeMTMC-VID-Attribute), achieving state-of-the-art results. The source code and dataset will be released on https://github.com/Event-AHU/OpenPAR

Fangyu Ding, Haiyang Wang, Zhixuan Chu et al.

Graph invariant learning (GIL) has been an effective approach to discovering the invariant relationships between graph data and its labels for different graph learning tasks under various distribution shifts. Many recent endeavors of GIL focus on extracting the invariant subgraph from the input graph for prediction as a regularization strategy to improve the generalization performance of graph learning. Despite their success, such methods also have various limitations in obtaining their invariant subgraphs. In this paper, we provide in-depth analyses of the drawbacks of existing works and propose corresponding principles of our invariant subgraph extraction: 1) the sparsity, to filter out the variant features, 2) the softness, for a broader solution space, and 3) the differentiability, for a soundly end-to-end optimization. To meet these principles in one shot, we leverage the Optimal Transport (OT) theory and propose a novel graph attention mechanism called Graph Sinkhorn Attention (GSINA). This novel approach serves as a powerful regularization method for GIL tasks. By GSINA, we are able to obtain meaningful, differentiable invariant subgraphs with controllable sparsity and softness. Moreover, GSINA is a general graph learning framework that could handle GIL tasks of multiple data grain levels. Extensive experiments on both synthetic and real-world datasets validate the superiority of our GSINA, which outperforms the state-of-the-art GIL methods by large margins on both graph-level tasks and node-level tasks. Our code is publicly available at \url{https://github.com/dingfangyu/GSINA}.

Haiyang Wang, Hao Tang, Li Jiang et al.

This paper proposes a simple, yet effective framework, called GiT, simultaneously applicable for various vision tasks only with a vanilla ViT. Motivated by the universality of the Multi-layer Transformer architecture (e.g, GPT) widely used in large language models (LLMs), we seek to broaden its scope to serve as a powerful vision foundation model (VFM). However, unlike language modeling, visual tasks typically require specific modules, such as bounding box heads for detection and pixel decoders for segmentation, greatly hindering the application of powerful multi-layer transformers in the vision domain. To solve this, we design a universal language interface that empowers the successful auto-regressive decoding to adeptly unify various visual tasks, from image-level understanding (e.g., captioning), over sparse perception (e.g., detection), to dense prediction (e.g., segmentation). Based on the above designs, the entire model is composed solely of a ViT, without any specific additions, offering a remarkable architectural simplification. GiT is a multi-task visual model, jointly trained across five representative benchmarks without task-specific fine-tuning. Interestingly, our GiT builds a new benchmark in generalist performance, and fosters mutual enhancement across tasks, leading to significant improvements compared to isolated training. This reflects a similar impact observed in LLMs. Further enriching training with 27 datasets, GiT achieves strong zero-shot results over various tasks. Due to its simple design, this paradigm holds promise for narrowing the architectural gap between vision and language. Code and models will be available at \url{https://github.com/Haiyang-W/GiT}.

Haiyang Wang, Guangyu Zhou, Siqi Liu et al.

Motivation: Predicting Drug-Target Interaction (DTI) is a well-studied topic in bioinformatics due to its relevance in the fields of proteomics and pharmaceutical research. Although many machine learning methods have been successfully applied in this task, few of them aim at leveraging the inherent heterogeneous graph structure in the DTI network to address the challenge. For better learning and interpreting the DTI topological structure and the similarity, it is desirable to have methods specifically for predicting interactions from the graph structure. Results: We present an end-to-end framework, DTI-GAT (Drug-Target Interaction prediction with Graph Attention networks) for DTI predictions. DTI-GAT incorporates a deep neural network architecture that operates on graph-structured data with the attention mechanism, which leverages both the interaction patterns and the features of drug and protein sequences. DTI-GAT facilitates the interpretation of the DTI topological structure by assigning different attention weights to each node with the self-attention mechanism. Experimental evaluations show that DTI-GAT outperforms various state-of-the-art systems on the binary DTI prediction problem. Moreover, the independent study results further demonstrate that our model can be generalized better than other conventional methods. Availability: The source code and all datasets are available at https://github.com/Haiyang-W/DTI-GRAPH

Yusong Lin, Haiyang Wang, Shuzhe Wu et al.

Agentic coding requires agents to effectively interact with runtime environments, e.g., command line interfaces (CLI), so as to complete tasks like resolving dependency issues, fixing system problems, etc. But it remains underexplored how such environment-intensive tasks can be obtained at scale to enhance agents' capabilities. To address this, based on an analogy between the Dockerfile and the agentic task, we propose to employ agents to simulate and explore environment histories, guided by execution feedback. By tracing histories of a healthy environment, its state can be inverted to an earlier one with runtime failures, from which a task can be derived by packing the buggy state and the corresponding error messages. With our method, named CLI-Gym, a total of 1,655 environment-intensive tasks are derived, being the largest collection of its kind. Moreover, with curated successful trajectories, our fine-tuned model, named LiberCoder, achieves substantial absolute improvements of +21.1% (to 46.1%) on Terminal-Bench, outperforming various strong baselines. To our knowledge, this is the first public pipeline for scalable derivation of environment-intensive tasks.

Yue Zhang, Zhiliang Tian, Shicheng Zhou et al.

Legal Judgment Prediction (LJP) is a pivotal task in legal AI. Existing semantic-enhanced LJP models integrate judicial precedents and legal knowledge for high performance. But they neglect legal reasoning logic, a critical component of legal judgments requiring rigorous logical analysis. Although some approaches utilize legal reasoning logic for high-quality predictions, their logic rigidity hinders adaptation to case-specific logical frameworks, particularly in complex cases that are lengthy and detailed. This paper proposes a rule-enhanced legal judgment prediction framework based on first-order logic (FOL) formalism and comparative learning (CL) to develop an adaptive adjustment mechanism for legal judgment logic and further enhance performance in LJP. Inspired by the process of human exam preparation, our method follows a three-stage approach: first, we initialize judgment rules using the FOL formalism to capture complex reasoning logic accurately; next, we propose a Confusion-aware Contrastive Learning (CACL) to dynamically optimize the judgment rules through a quiz consisting of confusable cases; finally, we utilize the optimized judgment rules to predict legal judgments. Experimental results on two public datasets show superior performance across all metrics. The code is publicly available{https://anonymous.4open.science/r/RLJP-FDF1}.

Ming Wang, Fang Wang, Minghao Hu et al.

Long-form article generation (LFAG) presents challenges such as maintaining logical consistency, comprehensive topic coverage, and narrative coherence across extended articles. Existing datasets often lack both the hierarchical structure and fine-grained annotation needed to effectively decompose tasks, resulting in shallow, disorganized article generation. To address these limitations, we introduce DeFine, a Decomposed and Fine-grained annotated dataset for long-form article generation. DeFine is characterized by its hierarchical decomposition strategy and the integration of domain-specific knowledge with multi-level annotations, ensuring granular control and enhanced depth in article generation. To construct the dataset, a multi-agent collaborative pipeline is proposed, which systematically segments the generation process into four parts: Data Miner, Cite Retreiver, Q&A Annotator and Data Cleaner. To validate the effectiveness of DeFine, we designed and tested three LFAG baselines: the web retrieval, the local retrieval, and the grounded reference. We fine-tuned the Qwen2-7b-Instruct model using the DeFine training dataset. The experimental results showed significant improvements in text quality, specifically in topic coverage, depth of information, and content fidelity. Our dataset publicly available to facilitate future research.

Yuan Zhou, Haiyang Wang, Shuwei Huo et al.

Self-attention architectures have emerged as a recent advancement for improving the performance of vision tasks. Manual determination of the architecture for self-attention networks relies on the experience of experts and cannot automatically adapt to various scenarios. Meanwhile, neural architecture search (NAS) has significantly advanced the automatic design of neural architectures. Thus, it is appropriate to consider using NAS methods to discover a better self-attention architecture automatically. However, it is challenging to directly use existing NAS methods to search attention networks because of the uniform cell-based search space and the lack of long-term content dependencies. To address this issue, we propose a full-attention based NAS method. More specifically, a stage-wise search space is constructed that allows various attention operations to be adopted for different layers of a network. To extract global features, a self-supervised search algorithm is proposed that uses context auto-regression to discover the full-attention architecture. To verify the efficacy of the proposed methods, we conducted extensive experiments on various learning tasks, including image classification, fine-grained image recognition, and zero-shot image retrieval. The empirical results show strong evidence that our method is capable of discovering high-performance, full-attention architectures while guaranteeing the required search efficiency.

Jikai Jin, Bohang Zhang, Haiyang Wang et al.

Distributionally robust optimization (DRO) is a widely-used approach to learn models that are robust against distribution shift. Compared with the standard optimization setting, the objective function in DRO is more difficult to optimize, and most of the existing theoretical results make strong assumptions on the loss function. In this work we bridge the gap by studying DRO algorithms for general smooth non-convex losses. By carefully exploiting the specific form of the DRO objective, we are able to provide non-asymptotic convergence guarantees even though the objective function is possibly non-convex, non-smooth and has unbounded gradient noise. In particular, we prove that a special algorithm called the mini-batch normalized gradient descent with momentum, can find an $ε$ first-order stationary point within $O( ε^{-4} )$ gradient complexity. We also discuss the conditional value-at-risk (CVaR) setting, where we propose a penalized DRO objective based on a smoothed version of the CVaR that allows us to obtain a similar convergence guarantee. We finally verify our theoretical results in a number of tasks and find that the proposed algorithm can consistently achieve prominent acceleration.

Haiyang Wang, Wenguan Wang, Xizhou Zhu et al.

As a fundamental problem for Artificial Intelligence, multi-agent system (MAS) is making rapid progress, mainly driven by multi-agent reinforcement learning (MARL) techniques. However, previous MARL methods largely focused on grid-world like or game environments; MAS in visually rich environments has remained less explored. To narrow this gap and emphasize the crucial role of perception in MAS, we propose a large-scale 3D dataset, CollaVN, for multi-agent visual navigation (MAVN). In CollaVN, multiple agents are entailed to cooperatively navigate across photo-realistic environments to reach target locations. Diverse MAVN variants are explored to make our problem more general. Moreover, a memory-augmented communication framework is proposed. Each agent is equipped with a private, external memory to persistently store communication information. This allows agents to make better use of their past communication information, enabling more efficient collaboration and robust long-term planning. In our experiments, several baselines and evaluation metrics are designed. We also empirically verify the efficacy of our proposed MARL approach across different MAVN task settings.

Longyuan Li, Junchi Yan, Haiyang Wang et al.

Deep generative models have demonstrated their effectiveness in learning latent representation and modeling complex dependencies of time series. In this paper, we present a Smoothness-Inducing Sequential Variational Auto-Encoder (SISVAE) model for robust estimation and anomaly detection of multi-dimensional time series. Our model is based on Variational Auto-Encoder (VAE), and its backbone is fulfilled by a Recurrent Neural Network to capture latent temporal structures of time series for both generative model and inference model. Specifically, our model parameterizes mean and variance for each time-stamp with flexible neural networks, resulting in a non-stationary model that can work without the assumption of constant noise as commonly made by existing Markov models. However, such a flexibility may cause the model fragile to anomalies. To achieve robust density estimation which can also benefit detection tasks, we propose a smoothness-inducing prior over possible estimations. The proposed prior works as a regularizer that places penalty at non-smooth reconstructions. Our model is learned efficiently with a novel stochastic gradient variational Bayes estimator. In particular, we study two decision criteria for anomaly detection: reconstruction probability and reconstruction error. We show the effectiveness of our model on both synthetic datasets and public real-world benchmarks.

Wenqiang Xu, Haiyang Wang, Fubo Qi et al.

In this paper, we propose a novel top-down instance segmentation framework based on explicit shape encoding, named \textbf{ESE-Seg}. It largely reduces the computational consumption of the instance segmentation by explicitly decoding the multiple object shapes with tensor operations, thus performs the instance segmentation at almost the same speed as the object detection. ESE-Seg is based on a novel shape signature Inner-center Radius (IR), Chebyshev polynomial fitting and the strong modern object detectors. ESE-Seg with YOLOv3 outperforms the Mask R-CNN on Pascal VOC 2012 at mAP$^r$@0.5 while 7 times faster.

Yutong Xie, Haiyang Wang, Yan Hao et al.

In this paper, we propose a data-driven visual rhythm prediction method, which overcomes the previous works' deficiency that predictions are made primarily by human-crafted hard rules. In our approach, we first extract features including original frames and their residuals, optical flow, scene change, and body pose. These visual features will be next taken into an end-to-end neural network as inputs. Here we observe that there are some slight misaligning between features over the timeline and assume that this is due to the distinctions between how different features are computed. To solve this problem, the extracted features are aligned by an elaborately designed layer, which can also be applied to other models suffering from mismatched features, and boost performance. Then these aligned features are fed into sequence labeling layers implemented with BiLSTM and CRF to predict the onsets. Due to the lack of existing public training and evaluation set, we experiment on a dataset constructed by ourselves based on professionally edited Music Videos (MVs), and the F1 score of our approach reaches 79.6.

Haiyang Wang, Yong Tang, Ziyang Jia et al.

Recent researches have shown that deep forest ensemble achieves a considerable increase in classification accuracy compared with the general ensemble learning methods, especially when the training set is small. In this paper, we take advantage of deep forest ensemble and introduce the Dense Adaptive Cascade Forest (daForest). Our model has a better performance than the original Cascade Forest with three major features: first, we apply SAMME.R boosting algorithm to improve the performance of the model. It guarantees the improvement as the number of layers increases. Second, our model connects each layer to the subsequent ones in a feed-forward fashion, which enhances the capability of the model to resist performance degeneration. Third, we add a hyper-parameters optimization layer before the first classification layer, making our model spend less time to set up and find the optimal hyper-parameters. Experimental results show that daForest performs significantly well, and in some cases, even outperforms neural networks and achieves state-of-the-art results.

Honglun Zhang, Haiyang Wang, Xiaming Chen et al.

P2P lending presents as an innovative and flexible alternative for conventional lending institutions like banks, where lenders and borrowers directly make transactions and benefit each other without complicated verifications. However, due to lack of specialized laws, delegated monitoring and effective managements, P2P platforms may spawn potential risks, such as withdraw failures, investigation involvements and even runaway bosses, which cause great losses to lenders and are especially serious and notorious in China. Although there are abundant public information and data available on the Internet related to P2P platforms, challenges of multi-sourcing and heterogeneity matter. In this paper, we promote a novel deep learning model, OMNIRank, which comprehends multi-dimensional features of P2P platforms for risk quantification and produces scores for ranking. We first construct a large-scale flexible crawling framework and obtain great amounts of multi-source heterogeneous data of domestic P2P platforms since 2007 from the Internet. Purifications like duplication and noise removal, null handing, format unification and fusion are applied to improve data qualities. Then we extract deep features of P2P platforms via text comprehension, topic modeling, knowledge graph and sentiment analysis, which are delivered as inputs to OMNIRank, a deep learning model for risk quantification of P2P platforms. Finally, according to rankings generated by OMNIRank, we conduct flourish data visualizations and interactions, providing lenders with comprehensive information supports, decision suggestions and safety guarantees.

Cong Zhang, Jiangchuan Liu, Haiyang Wang

Crowdsourced Live Streaming (CLS), most notably Twitch.tv, has seen explosive growth in its popularity in the past few years. In such systems, any user can lively broadcast video content of interest to others, e.g., from a game player to many online viewers. To fulfill the demands from both massive and heterogeneous broadcasters and viewers, expensive server clusters have been deployed to provide video ingesting and transcoding services. Despite the existence of highly popular channels, a significant portion of the channels is indeed unpopular. Yet as our measurement shows, these broadcasters are consuming considerable system resources; in particular, 25% (resp. 30%) of bandwidth (resp. computation) resources are used by the broadcasters who do not have any viewers at all. In this paper, we closely examine the challenge of handling unpopular live-broadcasting channels in CLS systems and present a comprehensive solution for service partitioning on hybrid cloud. The trace-driven evaluation shows that our hybrid cloud-assisted design can smartly assign ingesting and transcoding tasks to the elastic cloud virtual machines, providing flexible system deployment cost-effectively.