Haoyu Chen

Haoze Sun, Wenbo Li, Jianzhuang Liu et al.

Existing super-resolution (SR) models primarily focus on restoring local texture details, often neglecting the global semantic information within the scene. This oversight can lead to the omission of crucial semantic details or the introduction of inaccurate textures during the recovery process. In our work, we introduce the Cognitive Super-Resolution (CoSeR) framework, empowering SR models with the capacity to comprehend low-resolution images. We achieve this by marrying image appearance and language understanding to generate a cognitive embedding, which not only activates prior information from large text-to-image diffusion models but also facilitates the generation of high-quality reference images to optimize the SR process. To further improve image fidelity, we propose a novel condition injection scheme called "All-in-Attention", consolidating all conditional information into a single module. Consequently, our method successfully restores semantically correct and photorealistic details, demonstrating state-of-the-art performance across multiple benchmarks. Code: https://github.com/VINHYU/CoSeR

Cursor Research, Aaron Chan, Ahmed Shalaby et al. · berkeley, microsoft-research

Composer 2 is a specialized model designed for agentic software engineering. The model demonstrates strong long-term planning and coding intelligence while maintaining the ability to efficiently solve problems for interactive use. The model is trained in two phases: first, continued pretraining to improve the model's knowledge and latent coding ability, followed by large-scale reinforcement learning to improve end-to-end coding performance through stronger reasoning, accurate multi-step execution, and coherence on long-horizon realistic coding problems. We develop infrastructure to support training in the same Cursor harness that is used by the deployed model, with equivalent tools and structure, and use environments that match real problems closely. To measure the ability of the model on increasingly difficult tasks, we introduce a benchmark derived from real software engineering problems in large codebases including our own. Composer 2 is a frontier-level coding model and demonstrates a process for training strong domain-specialized models. On our CursorBench evaluations the model achieves a major improvement in accuracy compared to previous Composer models (61.3). On public benchmarks the model scores 61.7 on Terminal-Bench and 73.7 on SWE-bench Multilingual in our harness, comparable to state-of-the-art systems.

Meng Wang, Tian Lin, Lianyu Wang et al.

Failure to recognize samples from the classes unseen during training is a major limitation of artificial intelligence in the real-world implementation for recognition and classification of retinal anomalies. We established an uncertainty-inspired open-set (UIOS) model, which was trained with fundus images of 9 retinal conditions. Besides assessing the probability of each category, UIOS also calculated an uncertainty score to express its confidence. Our UIOS model with thresholding strategy achieved an F1 score of 99.55%, 97.01% and 91.91% for the internal testing set, external target categories (TC)-JSIEC dataset and TC-unseen testing set, respectively, compared to the F1 score of 92.20%, 80.69% and 64.74% by the standard AI model. Furthermore, UIOS correctly predicted high uncertainty scores, which would prompt the need for a manual check in the datasets of non-target categories retinal diseases, low-quality fundus images, and non-fundus images. UIOS provides a robust method for real-world screening of retinal anomalies.

Haoyu Chen, Jinjin Gu, Yihao Liu et al.

When capturing and storing images, devices inevitably introduce noise. Reducing this noise is a critical task called image denoising. Deep learning has become the de facto method for image denoising, especially with the emergence of Transformer-based models that have achieved notable state-of-the-art results on various image tasks. However, deep learning-based methods often suffer from a lack of generalization ability. For example, deep models trained on Gaussian noise may perform poorly when tested on other noise distributions. To address this issue, we present a novel approach to enhance the generalization performance of denoising networks, known as masked training. Our method involves masking random pixels of the input image and reconstructing the missing information during training. We also mask out the features in the self-attention layers to avoid the impact of training-testing inconsistency. Our approach exhibits better generalization ability than other deep learning models and is directly applicable to real-world scenarios. Additionally, our interpretability analysis demonstrates the superiority of our method.

Paul Baltescu, Haoyu Chen, Nikil Pancha et al.

Learned embeddings for products are an important building block for web-scale e-commerce recommendation systems. At Pinterest, we build a single set of product embeddings called ItemSage to provide relevant recommendations in all shopping use cases including user, image and search based recommendations. This approach has led to significant improvements in engagement and conversion metrics, while reducing both infrastructure and maintenance cost. While most prior work focuses on building product embeddings from features coming from a single modality, we introduce a transformer-based architecture capable of aggregating information from both text and image modalities and show that it significantly outperforms single modality baselines. We also utilize multi-task learning to make ItemSage optimized for several engagement types, leading to a candidate generation system that is efficient for all of the engagement objectives of the end-to-end recommendation system. Extensive offline experiments are conducted to illustrate the effectiveness of our approach and results from online A/B experiments show substantial gains in key business metrics (up to +7% gross merchandise value/user and +11% click volume).

Tu Trinh, Haoyu Chen, Daniel S. Brown · berkeley

We examine the problem of determining demonstration sufficiency: how can a robot self-assess whether it has received enough demonstrations from an expert to ensure a desired level of performance? To address this problem, we propose a novel self-assessment approach based on Bayesian inverse reinforcement learning and value-at-risk, enabling learning-from-demonstration ("LfD") robots to compute high-confidence bounds on their performance and use these bounds to determine when they have a sufficient number of demonstrations. We propose and evaluate two definitions of sufficiency: (1) normalized expected value difference, which measures regret with respect to the human's unobserved reward function, and (2) percent improvement over a baseline policy. We demonstrate how to formulate high-confidence bounds on both of these metrics. We evaluate our approach in simulation for both discrete and continuous state-space domains and illustrate the feasibility of developing a robotic system that can accurately evaluate demonstration sufficiency. We also show that the robot can utilize active learning in asking for demonstrations from specific states which results in fewer demos needed for the robot to still maintain high confidence in its policy. Finally, via a user study, we show that our approach successfully enables robots to perform at users' desired performance levels, without needing too many or perfectly optimal demonstrations.

Haotian Liu, Haoyu Chen, Chenhui Pan et al.

Face forgery detection encompasses multiple critical tasks, including identifying forged images and videos and localizing manipulated regions and temporal segments. Current approaches typically employ task-specific models with independent architectures, leading to computational redundancy and ignoring potential correlations across related tasks. We introduce OmniFD, a unified framework that jointly addresses four core face forgery detection tasks within a single model, i.e., image and video classification, spatial localization, and temporal localization. Our architecture consists of three principal components: (1) a shared Swin Transformer encoder that extracts unified 4D spatiotemporal representations from both images and video inputs, (2) a cross-task interaction module with learnable queries that dynamically captures inter-task dependencies through attention-based reasoning, and (3) lightweight decoding heads that transform refined representations into corresponding predictions for all FFD tasks. Extensive experiments demonstrate OmniFD's advantage over task-specific models. Its unified design leverages multi-task learning to capture generalized representations across tasks, especially enabling fine-grained knowledge transfer that facilitates other tasks. For example, video classification accuracy improves by 4.63% when image data are incorporated. Furthermore, by unifying images, videos and the four tasks within one framework, OmniFD achieves superior performance across diverse benchmarks with high efficiency and scalability, e.g., reducing 63% model parameters and 50% training time. It establishes a practical and generalizable solution for comprehensive face forgery detection in real-world applications. The source code is made available at https://github.com/haotianll/OmniFD.

Haoyu Chen, Linqi Song, Zhenxing Qian et al.

Data hiding with deep neural networks (DNNs) has experienced impressive successes in recent years. A prevailing scheme is to train an autoencoder, consisting of an encoding network to embed (or transform) secret messages in (or into) a carrier, and a decoding network to extract the hidden messages. This scheme may suffer from several limitations regarding practicability, security, and embedding capacity. In this work, we describe a different computational framework to hide images in deep probabilistic models. Specifically, we use a DNN to model the probability density of cover images, and hide a secret image in one particular location of the learned distribution. As an instantiation, we adopt a SinGAN, a pyramid of generative adversarial networks (GANs), to learn the patch distribution of one cover image. We hide the secret image by fitting a deterministic mapping from a fixed set of noise maps (generated by an embedding key) to the secret image during patch distribution learning. The stego SinGAN, behaving as the original SinGAN, is publicly communicated; only the receiver with the embedding key is able to extract the secret image. We demonstrate the feasibility of our SinGAN approach in terms of extraction accuracy and model security. Moreover, we show the flexibility of the proposed method in terms of hiding multiple images for different receivers and obfuscating the secret image.

Jinsheng Wei, Haoyu Chen, Guanming Lu et al.

Micro-expression recognition (MER) draws intensive research interest as micro-expressions (MEs) can infer genuine emotions. Prior information can guide the model to learn discriminative ME features effectively. However, most works focus on researching the general models with a stronger representation ability to adaptively aggregate ME movement information in a holistic way, which may ignore the prior information and properties of MEs. To solve this issue, driven by the prior information that the category of ME can be inferred by the relationship between the actions of facial different components, this work designs a novel model that can conform to this prior information and learn ME movement features in an interpretable way. Specifically, this paper proposes a Decomposition and Reconstruction-based Graph Representation Learning (DeRe-GRL) model to effectively learn high-level ME features. DeRe-GRL includes two modules: Action Decomposition Module (ADM) and Relation Reconstruction Module (RRM), where ADM learns action features of facial key components and RRM explores the relationship between these action features. Based on facial key components, ADM divides the geometric movement features extracted by the graph model-based backbone into several sub-features, and learns the map matrix to map these sub-features into multiple action features; then, RRM learns weights to weight all action features to build the relationship between action features. The experimental results demonstrate the effectiveness of the proposed modules, and the proposed method achieves competitive performance.

Ke Zou, Tian Lin, Xuedong Yuan et al.

Multimodality eye disease screening is crucial in ophthalmology as it integrates information from diverse sources to complement their respective performances. However, the existing methods are weak in assessing the reliability of each unimodality, and directly fusing an unreliable modality may cause screening errors. To address this issue, we introduce a novel multimodality evidential fusion pipeline for eye disease screening, EyeMoSt, which provides a measure of confidence for unimodality and elegantly integrates the multimodality information from a multi-distribution fusion perspective. Specifically, our model estimates both local uncertainty for unimodality and global uncertainty for the fusion modality to produce reliable classification results. More importantly, the proposed mixture of Student's $t$ distributions adaptively integrates different modalities to endow the model with heavy-tailed properties, increasing robustness and reliability. Our experimental findings on both public and in-house datasets show that our model is more reliable than current methods. Additionally, EyeMost has the potential ability to serve as a data quality discriminator, enabling reliable decision-making for multimodality eye disease screening.

Haoyu Chen, Zhihua Wang, Yang Yang et al.

Most well-established and widely used color difference (CD) metrics are handcrafted and subject-calibrated against uniformly colored patches, which do not generalize well to photographic images characterized by natural scene complexities. Constructing CD formulae for photographic images is still an active research topic in imaging/illumination, vision science, and color science communities. In this paper, we aim to learn a deep CD metric for photographic images with four desirable properties. First, it well aligns with the observations in vision science that color and form are linked inextricably in visual cortical processing. Second, it is a proper metric in the mathematical sense. Third, it computes accurate CDs between photographic images, differing mainly in color appearances. Fourth, it is robust to mild geometric distortions (e.g., translation or due to parallax), which are often present in photographic images of the same scene captured by different digital cameras. We show that all these properties can be satisfied at once by learning a multi-scale autoregressive normalizing flow for feature transform, followed by the Euclidean distance which is linearly proportional to the human perceptual CD. Quantitative and qualitative experiments on the large-scale SPCD dataset demonstrate the promise of the learned CD metric.

Sofoklis Kakouros, Fang Kang, Haoyu Chen

This work presents iMiGUE-Speech, an extension of the iMiGUE dataset that provides a spontaneous affective corpus for studying emotional and affective states. The new release focuses on speech and enriches the original dataset with additional metadata, including speech transcripts, speaker-role separation between interviewer and interviewee, and word-level forced alignments. Unlike existing emotional speech datasets that rely on acted or laboratory-elicited emotions, iMiGUE-Speech captures spontaneous affect arising naturally from real match outcomes. To demonstrate the utility of the dataset and establish initial benchmarks, we introduce two evaluation tasks for comparative assessment: speech emotion recognition and transcript-based sentiment analysis. These tasks leverage state-of-the-art pre-trained representations to assess the dataset's ability to capture spontaneous affective states from both acoustic and linguistic modalities. iMiGUE-Speech can also be synchronously paired with micro-gesture annotations from the original iMiGUE dataset, forming a uniquely multimodal resource for studying speech-gesture affective dynamics. The extended dataset is available at https://github.com/CV-AC/imigue-speech.

Jingjing Ren, Wenbo Li, Haoyu Chen et al.

Ultra-high-resolution image generation poses great challenges, such as increased semantic planning complexity and detail synthesis difficulties, alongside substantial training resource demands. We present UltraPixel, a novel architecture utilizing cascade diffusion models to generate high-quality images at multiple resolutions (\textit{e.g.}, 1K to 6K) within a single model, while maintaining computational efficiency. UltraPixel leverages semantics-rich representations of lower-resolution images in the later denoising stage to guide the whole generation of highly detailed high-resolution images, significantly reducing complexity. Furthermore, we introduce implicit neural representations for continuous upsampling and scale-aware normalization layers adaptable to various resolutions. Notably, both low- and high-resolution processes are performed in the most compact space, sharing the majority of parameters with less than 3$\%$ additional parameters for high-resolution outputs, largely enhancing training and inference efficiency. Our model achieves fast training with reduced data requirements, producing photo-realistic high-resolution images and demonstrating state-of-the-art performance in extensive experiments.

Haoyu Chen, Wenbo Li, Jinjin Gu et al.

Natural images captured by mobile devices often suffer from multiple types of degradation, such as noise, blur, and low light. Traditional image restoration methods require manual selection of specific tasks, algorithms, and execution sequences, which is time-consuming and may yield suboptimal results. All-in-one models, though capable of handling multiple tasks, typically support only a limited range and often produce overly smooth, low-fidelity outcomes due to their broad data distribution fitting. To address these challenges, we first define a new pipeline for restoring images with multiple degradations, and then introduce RestoreAgent, an intelligent image restoration system leveraging multimodal large language models. RestoreAgent autonomously assesses the type and extent of degradation in input images and performs restoration through (1) determining the appropriate restoration tasks, (2) optimizing the task sequence, (3) selecting the most suitable models, and (4) executing the restoration. Experimental results demonstrate the superior performance of RestoreAgent in handling complex degradation, surpassing human experts. Furthermore, the system modular design facilitates the fast integration of new tasks and models, enhancing its flexibility and scalability for various applications.

Yawen Cui, Wanxia Deng, Haoyu Chen et al.

Given a model well-trained with a large-scale base dataset, Few-Shot Class-Incremental Learning (FSCIL) aims at incrementally learning novel classes from a few labeled samples by avoiding overfitting, without catastrophically forgetting all encountered classes previously. Currently, semi-supervised learning technique that harnesses freely-available unlabeled data to compensate for limited labeled data can boost the performance in numerous vision tasks, which heuristically can be applied to tackle issues in FSCIL, i.e., the Semi-supervised FSCIL (Semi-FSCIL). So far, very limited work focuses on the Semi-FSCIL task, leaving the adaptability issue of semi-supervised learning to the FSCIL task unresolved. In this paper, we focus on this adaptability issue and present a simple yet efficient Semi-FSCIL framework named Uncertainty-aware Distillation with Class-Equilibrium (UaD-CE), encompassing two modules UaD and CE. Specifically, when incorporating unlabeled data into each incremental session, we introduce the CE module that employs a class-balanced self-training to avoid the gradual dominance of easy-to-classified classes on pseudo-label generation. To distill reliable knowledge from the reference model, we further implement the UaD module that combines uncertainty-guided knowledge refinement with adaptive distillation. Comprehensive experiments on three benchmark datasets demonstrate that our method can boost the adaptability of unlabeled data with the semi-supervised learning technique in FSCIL tasks.

Qiujing Lu, Weiqiao Han, Jeffrey Ling et al.

Predicting future trajectories of road agents is a critical task for autonomous driving. Recent goal-based trajectory prediction methods, such as DenseTNT and PECNet, have shown good performance on prediction tasks on public datasets. However, they usually require complicated goal-selection algorithms and optimization. In this work, we propose KEMP, a hierarchical end-to-end deep learning framework for trajectory prediction. At the core of our framework is keyframe-based trajectory prediction, where keyframes are representative states that trace out the general direction of the trajectory. KEMP first predicts keyframes conditioned on the road context, and then fills in intermediate states conditioned on the keyframes and the road context. Under our general framework, goal-conditioned methods are special cases in which the number of keyframes equal to one. Unlike goal-conditioned methods, our keyframe predictor is learned automatically and does not require hand-crafted goal-selection algorithms. We evaluate our model on public benchmarks and our model ranked 1st on Waymo Open Motion Dataset Leaderboard (as of September 1, 2021).

Xiaocheng Tang, Soheil Sadeghi Eshkevari, Haoyu Chen et al.

Transformers have enabled breakthroughs in NLP and computer vision, and have recently began to show promising performance in trajectory prediction for Autonomous Vehicle (AV). How to efficiently model the interactive relationships between the ego agent and other road and dynamic objects remains challenging for the standard attention module. In this work we propose a general Transformer-like architectural module MnM network equipped with novel masked goal conditioning training procedures for AV trajectory prediction. The resulted model, named golfer, achieves state-of-the-art performance, winning the 2nd place in the 2022 Waymo Open Dataset Motion Prediction Challenge and ranked 1st place according to minADE.

Wufei Ma, Haoyu Chen, Guofeng Zhang et al.

3D spatial reasoning is the ability to analyze and interpret the positions, orientations, and spatial relationships of objects within the 3D space. This allows models to develop a comprehensive understanding of the 3D scene, enabling their applicability to a broader range of areas, such as autonomous navigation, robotics, and AR/VR. While large multi-modal models (LMMs) have achieved remarkable progress in a wide range of image and video understanding tasks, their capabilities to perform 3D spatial reasoning on diverse natural images are less studied. In this work we present the first comprehensive 3D spatial reasoning benchmark, 3DSRBench, with 2,772 manually annotated visual question-answer pairs across 12 question types. We conduct robust and thorough evaluation of 3D spatial reasoning abilities by balancing data distribution and adopting a novel FlipEval strategy. To further study the robustness of 3D spatial reasoning w.r.t. camera 3D viewpoints, our 3DSRBench includes two subsets with 3D spatial reasoning questions on paired images with common and uncommon viewpoints. We benchmark a wide range of open-sourced and proprietary LMMs, uncovering their limitations in various aspects of 3D awareness, such as height, orientation, location, and multi-object reasoning, as well as their degraded performance on images from uncommon 6D viewpoints. Our 3DSRBench provide valuable findings and insights about future development of LMMs with strong spatial reasoning abilities. Our project page is available at https://3dsrbench.github.io/.

SiXiang Chen, Jianyu Lai, Jialin Gao et al.

Generating aesthetic posters is more challenging than simple design images: it requires not only precise text rendering but also the seamless integration of abstract artistic content, striking layouts, and overall stylistic harmony. To address this, we propose PosterCraft, a unified framework that abandons prior modular pipelines and rigid, predefined layouts, allowing the model to freely explore coherent, visually compelling compositions. PosterCraft employs a carefully designed, cascaded workflow to optimize the generation of high-aesthetic posters: (i) large-scale text-rendering optimization on our newly introduced Text-Render-2M dataset; (ii) region-aware supervised fine-tuning on HQ-Poster100K; (iii) aesthetic-text-reinforcement learning via best-of-n preference optimization; and (iv) joint vision-language feedback refinement. Each stage is supported by a fully automated data-construction pipeline tailored to its specific needs, enabling robust training without complex architectural modifications. Evaluated on multiple experiments, PosterCraft significantly outperforms open-source baselines in rendering accuracy, layout coherence, and overall visual appeal-approaching the quality of SOTA commercial systems. Our code, models, and datasets can be found in the Project page: https://ephemeral182.github.io/PosterCraft

Alexander Vedernikov, Puneet Kumar, Haoyu Chen et al.

Engagement analysis finds various applications in healthcare, education, advertisement, services. Deep Neural Networks, used for analysis, possess complex architecture and need large amounts of input data, computational power, inference time. These constraints challenge embedding systems into devices for real-time use. To address these limitations, we present a novel two-stream feature fusion "Tensor-Convolution and Convolution-Transformer Network" (TCCT-Net) architecture. To better learn the meaningful patterns in the temporal-spatial domain, we design a "CT" stream that integrates a hybrid convolutional-transformer. In parallel, to efficiently extract rich patterns from the temporal-frequency domain and boost processing speed, we introduce a "TC" stream that uses Continuous Wavelet Transform (CWT) to represent information in a 2D tensor form. Evaluated on the EngageNet dataset, the proposed method outperforms existing baselines, utilizing only two behavioral features (head pose rotations) compared to the 98 used in baseline models. Furthermore, comparative analysis shows TCCT-Net's architecture offers an order-of-magnitude improvement in inference speed compared to state-of-the-art image-based Recurrent Neural Network (RNN) methods. The code will be released at https://github.com/vedernikovphoto/TCCT_Net.

Zuxian He, Xu Cheng, Zhaodong Sun et al.

Remote photoplethysmography (rPPG) is a non-contact technique that estimates physiological signals by analyzing subtle skin color changes in facial videos. Existing rPPG methods often encounter performance degradation under facial motion and occlusion scenarios due to their reliance on static and single-view facial videos. Thus, this work focuses on tackling the motion-induced occlusion problem for rPPG measurement in unconstrained multi-view facial videos. Specifically, we introduce a Multi-View rPPG Dataset (MVRD), a high-quality benchmark dataset featuring synchronized facial videos from three viewpoints under stationary, speaking, and head movement scenarios to better match real-world conditions. We also propose MVRD-rPPG, a unified multi-view rPPG learning framework that fuses complementary visual cues to maintain robust facial skin coverage, especially under motion conditions. Our method integrates an Adaptive Temporal Optical Compensation (ATOC) module for motion artifact suppression, a Rhythm-Visual Dual-Stream Network to disentangle rhythmic and appearance-related features, and a Multi-View Correlation-Aware Attention (MVCA) for adaptive view-wise signal aggregation. Furthermore, we introduce a Correlation Frequency Adversarial (CFA) learning strategy, which jointly enforces temporal accuracy, spectral consistency, and perceptual realism in the predicted signals. Extensive experiments and ablation studies on the MVRD dataset demonstrate the superiority of our approach. In the MVRD movement scenario, MVRD-rPPG achieves an MAE of 0.90 and a Pearson correlation coefficient (R) of 0.99. The source code and dataset will be made available.

Peibei Cao, Haoyu Chen, Jingzhe Ma et al.

High dynamic range (HDR) capture and display have seen significant growth in popularity driven by the advancements in technology and increasing consumer demand for superior image quality. As a result, HDR image compression is crucial to fully realize the benefits of HDR imaging without suffering from large file sizes and inefficient data handling. Conventionally, this is achieved by introducing a residual/gain map as additional metadata to bridge the gap between HDR and low dynamic range (LDR) images, making the former compatible with LDR image codecs but offering suboptimal rate-distortion performance. In this work, we initiate efforts towards end-to-end optimized HDR image compression for perceptually optimal storage and display. Specifically, we learn to compress an HDR image into two bitstreams: one for generating an LDR image to ensure compatibility with legacy LDR displays, and another as side information to aid HDR image reconstruction from the output LDR image. To measure the perceptual quality of output HDR and LDR images, we use two recently proposed image distortion metrics, both validated against human perceptual data of image quality and with reference to the uncompressed HDR image. Through end-to-end optimization for rate-distortion performance, our method dramatically improves HDR and LDR image quality at all bit rates.

Chengyan Wang, Haoyu Chen, Hui Wei et al.

Emotion understanding is a fundamental challenge in affective computing and artificial intelligence. While existing approaches predominantly focus on facial expressions and speech, they often overlook the rich emotional cues conveyed through body language. Recently, micro-gestures (MGs), unintentional, subconscious movements driven by inner feelings, have attracted increasing attention as an alternative to other cues. However, there are no existing large-scale datasets supporting the pre-training of the MG foundation model. To advance MG research, we present a new benchmark for micro-gesture-based emotion understanding, featuring key contributions with a novel dataset (iMiGUE-3K) and a series of foundation models for different tasks. Using a model-based crowd-sourcing data collection strategy, we construct iMiGUE-3K, the largest MG dataset to date. It comprises video recordings from 332 distinct professional tennis players' public press interviews over the past seven years, totaling more than 3.4K long video clips and 37 million frames. The dataset includes 32 micro-gesture classes with rich descriptive annotations, making it the first large-scale, in-the-wild, video dataset for fine-grained gesture-based emotion analysis. Built on iMiGUE-3K, we propose MG-FMs, a discriminative foundation model for transferable gesture presentation learning. Based on the foundation model, we establish five comprehensive evaluation tasks: MG recognition (unsupervised, semi-supervised, supervised), MG retrieval, and MG emotion recognition. Our systematic evaluation of representative methods demonstrates that micro-gesture-based analysis significantly improves emotion understanding. We hope this work can provide comprehensive tools for MG analysis and set a solid foundation for future research in psychological diagnostics, affective computing, and advanced human-computer interaction.

Jinsheng Wei, Fengzhou Guo, Yante Li et al.

Micro-expression (ME) action units (Micro-AUs) provide objective clues for fine-grained genuine emotion analysis. Most existing Micro-AU detection methods learn AU features from the whole facial image/video, which conflicts with the inherent locality of AU, resulting in insufficient perception of AU regions. In fact, each AU independently corresponds to specific localized facial muscle movements (local independence), while there is an inherent dependency between some AUs under specific emotional states (global dependency). Thus, this paper explores the effectiveness of the independence-to-dependency pattern and proposes a novel micro-AU detection framework, micro-AU CLIP, that uniquely decomposes the AU detection process into local semantic independence modeling (LSI) and global semantic dependency (GSD) modeling. In LSI, Patch Token Attention (PTA) is designed, mapping several local features within the AU region to the same feature space; In GSD, Global Dependency Attention (GDA) and Global Dependency Loss (GDLoss) are presented to model the global dependency relationships between different AUs, thereby enhancing each AU feature. Furthermore, considering CLIP's native limitations in micro-semantic alignment, a microAU contrastive loss (MiAUCL) is designed to learn AU features by a fine-grained alignment of visual and text features. Also, Micro-AU CLIP is effectively applied to ME recognition in an emotion-label-free way. The experimental results demonstrate that Micro-AU CLIP can fully learn fine-grained micro-AU features, achieving state-of-the-art performance.

Jinsheng Wei, Zhaodi Xu, Guanming Lu et al.

Micro-gesture recognition (MGR) is challenging due to subtle inter-class variations. Existing methods rely on category-level supervision, which is insufficient for capturing subtle and localized motion differences. Thus, this paper proposes a Fine-Grained Semantic Guidance Learning (FG-SGL) framework that jointly integrates fine-grained and category-level semantics to guide vision--language models in perceiving local MG motions. FG-SA adopts fine-grained semantic cues to guide the learning of local motion features, while CP-A enhances the separability of MG features through category-level semantic guidance. To support fine-grained semantic guidance, this work constructs a fine-grained textual dataset with human annotations that describes the dynamic process of MGs in four refined semantic dimensions. Furthermore, a Multi-Level Contrastive Optimization strategy is designed to jointly optimize both modules in a coarse-to-fine pattern. Experiments show that FG-SGL achieves competitive performance, validating the effectiveness of fine-grained semantic guidance for MGR.

Qingshan Hou, Yukun Zhou, Jocelyn Hui Lin Goh et al.

The advent of foundation models (FMs) is transforming medical domain. In ophthalmology, RETFound, a retina-specific FM pre-trained sequentially on 1.4 million natural images and 1.6 million retinal images, has demonstrated high adaptability across clinical applications. Conversely, DINOv2, a general-purpose vision FM pre-trained on 142 million natural images, has shown promise in non-medical domains. However, its applicability to clinical tasks remains underexplored. To address this, we conducted head-to-head evaluations by fine-tuning RETFound and three DINOv2 models (large, base, small) for ocular disease detection and systemic disease prediction tasks, across eight standardized open-source ocular datasets, as well as the Moorfields AlzEye and the UK Biobank datasets. DINOv2-large model outperformed RETFound in detecting diabetic retinopathy (AUROC=0.850-0.952 vs 0.823-0.944, across three datasets, all P<=0.007) and multi-class eye diseases (AUROC=0.892 vs. 0.846, P<0.001). In glaucoma, DINOv2-base model outperformed RETFound (AUROC=0.958 vs 0.940, P<0.001). Conversely, RETFound achieved superior performance over all DINOv2 models in predicting heart failure, myocardial infarction, and ischaemic stroke (AUROC=0.732-0.796 vs 0.663-0.771, all P<0.001). These trends persisted even with 10% of the fine-tuning data. These findings showcase the distinct scenarios where general-purpose and domain-specific FMs excel, highlighting the importance of aligning FM selection with task-specific requirements to optimise clinical performance.

Anirudhan Badrinath, Alex Yang, Kousik Rajesh et al. · stanford

Representation learning, a task of learning latent vectors to represent entities, is a key task in improving search and recommender systems in web applications. Various representation learning methods have been developed, including graph-based approaches for relationships among entities, sequence-based methods for capturing the temporal evolution of user activities, and content-based models for leveraging text and visual content. However, the development of a unifying framework that integrates these diverse techniques to support multiple applications remains a significant challenge. This paper presents OmniSage, a large-scale representation framework that learns universal representations for a variety of applications at Pinterest. OmniSage integrates graph neural networks with content-based models and user sequence models by employing multiple contrastive learning tasks to effectively process graph data, user sequence data, and content signals. To support the training and inference of OmniSage, we developed an efficient infrastructure capable of supporting Pinterest graphs with billions of nodes. The universal representations generated by OmniSage have significantly enhanced user experiences on Pinterest, leading to an approximate 2.5% increase in sitewide repins (saves) across five applications. This paper highlights the impact of unifying representation learning methods, and we make the model code publicly available at https://github.com/pinterest/atg-research/tree/main/omnisage.

Yueyi Yang, Haotian Liu, Fang Kang et al.

We explore the use of large language models (LLMs) for next-utterance prediction in human dialogue. Despite recent advances in LLMs demonstrating their ability to engage in natural conversations with users, we show that even leading models surprisingly struggle to predict a human speaker's next utterance. Instead, humans can readily anticipate forthcoming utterances based on multimodal cues, such as gestures, gaze, and emotional tone, from the context. To systematically examine whether LLMs can reproduce this ability, we propose SayNext-Bench, a benchmark that evaluates LLMs and Multimodal LLMs (MLLMs) on anticipating context-conditioned responses from multimodal cues spanning a variety of real-world scenarios. To support this benchmark, we build SayNext-PC, a novel large-scale dataset containing dialogues with rich multimodal cues. Building on this, we further develop a dual-route prediction MLLM, SayNext-Chat, that incorporates cognitively inspired design to emulate predictive processing in conversation. Experimental results demonstrate that our model outperforms state-of-the-art MLLMs in terms of lexical overlap, semantic similarity, and emotion consistency. Our results prove the feasibility of next-utterance prediction with LLMs from multimodal cues and emphasize the (i) indispensable role of multimodal cues and (ii) actively predictive processing as the foundation of natural human interaction, which is missing in current MLLMs. We hope that this exploration offers a new research entry toward more human-like, context-sensitive AI interaction for human-centered AI. Our benchmark and model can be accessed at https://saynext.github.io/.

Haoyu Chen, Xue Li, Kun Qian et al.

In Large Language Model (LLM) inference services, it is challenging to make a parallelism strategy configuration, to efficiently process the requests of variance context lengths. Requests of long context require high degree of parallelism to provide more memory for Key-Value (KV) Cache, while requests of short context prefer low degree of parallelism to increase concurrency, thus improving throughput. To maintain high throughput while supporting large context lengths on demand, we propose Amoeba, a runtime Tensor Parallel (TP) transformation for online LLM inference services, which adaptively adjusts the TP of running instances to align with the dynamics of incoming requests. Evaluations using real-world traces show that Amoeba improves throughput by 1.75x-6.57x compared to state-of-the-art solutions.

Licai Sun, Xingxun Jiang, Haoyu Chen et al.

Current facial emotion recognition systems are predominately trained to predict a fixed set of predefined categories or abstract dimensional values. This constrained form of supervision hinders generalization and applicability, as it reduces the rich and nuanced spectrum of emotions into oversimplified labels or scales. In contrast, natural language provides a more flexible, expressive, and interpretable way to represent emotions, offering a much broader source of supervision. Yet, leveraging semantically rich natural language captions as supervisory signals for facial emotion representation learning remains relatively underexplored, primarily due to two key challenges: 1) the lack of large-scale caption datasets with rich emotional semantics, and 2) the absence of effective frameworks tailored to harness such rich supervision. To this end, we introduce EmoCap100K, a large-scale facial emotion caption dataset comprising over 100,000 samples, featuring rich and structured semantic descriptions that capture both global affective states and fine-grained local facial behaviors. Building upon this dataset, we further propose EmoCapCLIP, which incorporates a joint global-local contrastive learning framework enhanced by a cross-modal guided positive mining module. This design facilitates the comprehensive exploitation of multi-level caption information while accommodating semantic similarities between closely related expressions. Extensive evaluations on over 20 benchmarks covering five tasks demonstrate the superior performance of our method, highlighting the promise of learning facial emotion representations from large-scale semantically rich captions. The code and data will be available at https://github.com/sunlicai/EmoCapCLIP.

Yan Jiang, Hao Yu, Mengting Wei et al.

Visible-infrared person re-identification (VI-ReID) is a challenging task that aims to match pedestrian images captured under varying lighting conditions, which has drawn intensive research attention and achieved promising results. However, existing methods adopt the centralized training, ignoring the potential privacy concerns as the data is distributed across multiple devices or entities in reality. In this paper, we propose L2RW+, a benchmark that brings VI-ReID closer to real-world applications. The core rationale behind L2RW+ is that incorporating decentralized training into VI-ReID can address privacy concerns in scenarios with limited data-sharing constrains. Specifically, we design protocols and corresponding algorithms for different privacy sensitivity levels. In our new benchmark, we simulate the training under real-world data conditions that: 1) data from each camera is completely isolated, or 2) different data entities (e.g., data controllers of a certain region) can selectively share the data. In this way, we simulate scenarios with strict privacy restrictions, which is closer to real-world conditions. Comprehensive experiments show the feasibility and potential of decentralized VI-ReID training at both image and video levels. In particular, with increasing data scales, the performance gap between decentralized and centralized training decreases, especially in video-level VI-ReID. In unseen domains, decentralized training even achieves performance comparable to SOTA centralized methods. This work offers a novel research entry for deploying VI-ReID into real-world scenarios and can benefit the community. Code is available at: https://github.com/Joey623/L2RW.

Weiyu Kong, Yun Lin, Xiwen Teoh et al.

Large Language Models (LLMs) have improved programming efficiency, but their performance degrades significantly as requirements scale; when faced with multi-modal documents containing hundreds of scenarios, LLMs often produce incorrect implementations or omit constraints. We propose Agentic Requirement Compilation (ARC), a technique that moves beyond simple code generation to requirement compilation, enabling the creation of runnable web systems directly from multi-modal DSL documents. ARC generates not only source code but also modular designs for UI, API, and database layers, enriched test suites (unit, modular, and integration), and detailed traceability for software maintenance. Our approach employs a bidirectional test-driven agentic loop: a top-down architecture phase decomposes requirements into verifiable interfaces, followed by a bottom-up implementation phase where agents generate code to satisfy those tests. ARC maintains strict traceability across requirements, design, and code to facilitate intelligent asset reuse. We evaluated ARC by generating six runnable web systems from documents spanning 50-200 multi-modal scenarios. Compared to state-of-the-art baselines, ARC-generated systems pass 50.6% more GUI tests on average. A user study with 21 participants showed that novice users can successfully write DSL documents for complex systems, such as a 10K-line ticket-booking system, in an average of 5.6 hours. These results demonstrate that ARC effectively transforms non-trivial requirement specifications into maintainable, runnable software.

Xiaoyu Yuan, Niklas Heikkala, Tiina Törmänen et al.

Understanding human mental states from natural behavior is crucial for intelligent systems in the real world. However, most current research focuses on predicting isolated mental state labels, lacking structured annotations of complex interpersonal interactions. To support structured analysis, we introduce MOTOR-Bench, a carefully-designed benchmark with a real-world dataset MOTOR-dataset, containing 1,440 multimodal video clips in collaborative learning scenarios, reflecting key real-world data challenges including natural class imbalance, visual noise, and domain-specific language. Each sample is labeled by educational experts based on self-regulated learning theory. We further evaluate several state-of-the-art multimodal large language models and multi-agent systems in a zero-shot setting on our MOTOR-Bench. However, their performance on this task remains limited, suggesting that existing methods still struggle with structured reasoning from observable behavior to deeper mental states. To address this challenge, we propose a reasoning multi-agent framework, named MOTOR-MAS. It coordinates multiple agents through a structured agent coordination mechanism to infer explicit behaviors, internal cognitions, and psychological emotions. Experimental results show that our MOTOR-MAS outperforms the best single-model benchmark by 15.93 points in Macro-F1 scores for the three labels of behavior, cognition, and emotion, and outperforms the general multi-agent benchmark by 10.2 points in internal cognition prediction.

Hao Yu, Haoyu Chen, Yan Jiang et al.

Self-attention (SA) has become the cornerstone of modern vision backbones for its powerful expressivity over traditional Convolutions (Conv). However, its quadratic complexity remains a critical bottleneck for practical applications. Given that Conv offers linear complexity and strong visual priors, continuing efforts have been made to promote the renaissance of Conv. However, a persistent performance chasm remains, highlighting that these modernizations have not yet captured the intrinsic expressivity that defines SA. In this paper, we re-examine the design of the CNNs, directed by a key question: what principles give SA its edge over Conv? As a result, we reveal two fundamental insights that challenge the long-standing design intuitions in prior research (e.g., Receptive field). The two findings are: (1) \textit{Adaptive routing}: SA dynamically regulates positional information flow according to semantic content, whereas Conv employs static kernels uniformly across all positions. (2) \textit{Lateral inhibition}: SA induces score competition among token weighting, effectively suppressing redundancy and sharpening representations, whereas Conv filters lack such inhibitory dynamics and exhibit considerable redundancy. Based on this, we propose \textit{Attentive Convolution} (ATConv), a principled reformulation of the convolutional operator that intrinsically injects these principles. Interestingly, with only $3\times3$ kernels, ATConv consistently outperforms various SA mechanisms in fundamental vision tasks. Building on ATConv, we introduce AttNet, a CNN family that can attain \textbf{84.4\%} ImageNet-1K Top-1 accuracy with only 27M parameters. In diffusion-based image generation, replacing all SA with the proposed $3\times 3$ ATConv in SiT-XL/2 reduces ImageNet FID by 0.15 in 400k steps with faster sampling. Code is available at: github.com/price112/Attentive-Convolution.

Zijian Wang, Wei Tong, Tingxuan Han et al.

Federated learning (FL) combined with local differential privacy (LDP) enables privacy-preserving model training across decentralized data sources. However, the decentralized data-management paradigm leaves LDPFL vulnerable to participants with malicious intent. The robustness of LDPFL protocols, particularly against model poisoning attacks (MPA), where adversaries inject malicious updates to disrupt global model convergence, remains insufficiently studied. In this paper, we propose a novel and extensible model poisoning attack framework tailored for LDPFL settings. Our approach is driven by the objective of maximizing the global training loss while adhering to local privacy constraints. To counter robust aggregation mechanisms such as Multi-Krum and trimmed mean, we develop adaptive attacks that embed carefully crafted constraints into a reverse training process, enabling evasion of these defenses. We evaluate our framework across three representative LDPFL protocols, three benchmark datasets, and two types of deep neural networks. Additionally, we investigate the influence of data heterogeneity and privacy budgets on attack effectiveness. Experimental results demonstrate that our adaptive attacks can significantly degrade the performance of the global model, revealing critical vulnerabilities and highlighting the need for more robust LDPFL defense strategies against MPA. Our code is available at https://github.com/ZiJW/LDPFL-Attack

Huihui Xu, Jiashi Lin, Haoyu Chen et al.

Referring Video Object Segmentation (RVOS) aims to segment out the object in a video referred by an expression. Current RVOS methods view referring expressions as unstructured sequences, neglecting their crucial semantic structure essential for referent reasoning. Besides, in contrast to image-referring expressions whose semantics focus only on object attributes and object-object relations, video-referring expressions also encompass event attributes and event-event temporal relations. This complexity challenges traditional structured reasoning image approaches. In this paper, we propose the Event Referential Reasoning (EventRR) framework. EventRR decouples RVOS into object summarization part and referent reasoning part. The summarization phase begins by summarizing each frame into a set of bottleneck tokens, which are then efficiently aggregated in the video-level summarization step to exchange the global cross-modal temporal context. For reasoning part, EventRR extracts semantic eventful structure of a video-referring expression into highly expressive Referential Event Graph (REG), which is a single-rooted directed acyclic graph. Guided by topological traversal of REG, we propose Temporal Concept-Role Reasoning (TCRR) to accumulate the referring score of each temporal query from REG leaf nodes to root node. Each reasoning step can be interpreted as a question-answer pair derived from the concept-role relations in REG. Extensive experiments across four widely recognized benchmark datasets, show that EventRR quantitatively and qualitatively outperforms state-of-the-art RVOS methods. Code is available at https://github.com/bio-mlhui/EventRR

Haoyu Chen, Hao Tang, Zitong Yu et al.

We present a customized 3D mesh Transformer model for the pose transfer task. As the 3D pose transfer essentially is a deformation procedure dependent on the given meshes, the intuition of this work is to perceive the geometric inconsistency between the given meshes with the powerful self-attention mechanism. Specifically, we propose a novel geometry-contrastive Transformer that has an efficient 3D structured perceiving ability to the global geometric inconsistencies across the given meshes. Moreover, locally, a simple yet efficient central geodesic contrastive loss is further proposed to improve the regional geometric-inconsistency learning. At last, we present a latent isometric regularization module together with a novel semi-synthesized dataset for the cross-dataset 3D pose transfer task towards unknown spaces. The massive experimental results prove the efficacy of our approach by showing state-of-the-art quantitative performances on SMPL-NPT, FAUST and our new proposed dataset SMG-3D datasets, as well as promising qualitative results on MG-cloth and SMAL datasets. It's demonstrated that our method can achieve robust 3D pose transfer and be generalized to challenging meshes from unknown spaces on cross-dataset tasks. The code and dataset are made available. Code is available: https://github.com/mikecheninoulu/CGT.

Haoyu Chen, Hao Tang, Nicu Sebe et al.

We present a novel task, i.e., animating a target 3D object through the motion of a raw driving sequence. In previous works, extra auxiliary correlations between source and target meshes or intermedia factors are inevitable to capture the motions in the driving sequences. Instead, we introduce AniFormer, a novel Transformer-based architecture, that generates animated 3D sequences by directly taking the raw driving sequences and arbitrary same-type target meshes as inputs. Specifically, we customize the Transformer architecture for 3D animation that generates mesh sequences by integrating styles from target meshes and motions from the driving meshes. Besides, instead of the conventional single regression head in the vanilla Transformer, AniFormer generates multiple frames as outputs to preserve the sequential consistency of the generated meshes. To achieve this, we carefully design a pair of regression constraints, i.e., motion and appearance constraints, that can provide strong regularization on the generated mesh sequences. Our AniFormer achieves high-fidelity, realistic, temporally coherent animated results and outperforms compared start-of-the-art methods on benchmarks of diverse categories. Code is available: https://github.com/mikecheninoulu/AniFormer.

Haoyu Chen, Hao Tang, Henglin Shi et al.

With the strength of deep generative models, 3D pose transfer regains intensive research interests in recent years. Existing methods mainly rely on a variety of constraints to achieve the pose transfer over 3D meshes, e.g., the need for manually encoding for shape and pose disentanglement. In this paper, we present an unsupervised approach to conduct the pose transfer between any arbitrate given 3D meshes. Specifically, a novel Intrinsic-Extrinsic Preserved Generative Adversarial Network (IEP-GAN) is presented for both intrinsic (i.e., shape) and extrinsic (i.e., pose) information preservation. Extrinsically, we propose a co-occurrence discriminator to capture the structural/pose invariance from distinct Laplacians of the mesh. Meanwhile, intrinsically, a local intrinsic-preserved loss is introduced to preserve the geodesic priors while avoiding heavy computations. At last, we show the possibility of using IEP-GAN to manipulate 3D human meshes in various ways, including pose transfer, identity swapping and pose interpolation with latent code vector arithmetic. The extensive experiments on various 3D datasets of humans, animals and hands qualitatively and quantitatively demonstrate the generality of our approach. Our proposed model produces better results and is substantially more efficient compared to recent state-of-the-art methods. Code is available: https://github.com/mikecheninoulu/Unsupervised_IEPGAN

Haoyu Chen, Jinjin Gu, Zhi Zhang

Convolutional neural networks have allowed remarkable advances in single image super-resolution (SISR) over the last decade. Among recent advances in SISR, attention mechanisms are crucial for high-performance SR models. However, the attention mechanism remains unclear on why and how it works in SISR. In this work, we attempt to quantify and visualize attention mechanisms in SISR and show that not all attention modules are equally beneficial. We then propose attention in attention network (A$^2$N) for more efficient and accurate SISR. Specifically, A$^2$N consists of a non-attention branch and a coupling attention branch. A dynamic attention module is proposed to generate weights for these two branches to suppress unwanted attention adjustments dynamically, where the weights change adaptively according to the input features. This allows attention modules to specialize to beneficial examples without otherwise penalties and thus greatly improve the capacity of the attention network with few parameters overhead. Experimental results demonstrate that our final model A$^2$N could achieve superior trade-off performances comparing with state-of-the-art networks of similar sizes. Codes are available at https://github.com/haoyuc/A2N.

Zitong Yu, Benjia Zhou, Jun Wan et al.

Gesture recognition has attracted considerable attention owing to its great potential in applications. Although the great progress has been made recently in multi-modal learning methods, existing methods still lack effective integration to fully explore synergies among spatio-temporal modalities effectively for gesture recognition. The problems are partially due to the fact that the existing manually designed network architectures have low efficiency in the joint learning of multi-modalities. In this paper, we propose the first neural architecture search (NAS)-based method for RGB-D gesture recognition. The proposed method includes two key components: 1) enhanced temporal representation via the proposed 3D Central Difference Convolution (3D-CDC) family, which is able to capture rich temporal context via aggregating temporal difference information; and 2) optimized backbones for multi-sampling-rate branches and lateral connections among varied modalities. The resultant multi-modal multi-rate network provides a new perspective to understand the relationship between RGB and depth modalities and their temporal dynamics. Comprehensive experiments are performed on three benchmark datasets (IsoGD, NvGesture, and EgoGesture), demonstrating the state-of-the-art performance in both single- and multi-modality settings.The code is available at https://github.com/ZitongYu/3DCDC-NAS

Kuofei Fang, Xinyi Che, Haomin Ouyang et al.

Embodied AI is a prominent research topic in both academia and industry. Current research centers on completing tasks based on explicit user instructions. However, for robots to integrate into human society, they must understand which actions are permissible and which are prohibited, even without explicit commands. We refer to the user-guided AI as passive intelligence and the unguided AI as active intelligence. This paper introduces RobotEQ, the first benchmark for active intelligence, aiming to assess whether existing models can comprehend and adhere to social norms in embodied scenarios. First, we construct RobotEQ-Data, a dataset consisting of 1,900 egocentric images, spanning 10 representative embodied categories and 56 subcategories. Through extensive manual annotation, we provide 5,353 action judgment questions and 1,286 spatial grounding questions, specifying appropriate robot actions across diverse scenarios. Furthermore, we establish RobotEQ-Bench to evaluate the performance of state-of-the-art models on this task. Experimental results show that current models still fall short in achieving reliable active intelligence, particularly in spatial grounding. Meanwhile, we observe that leveraging RAG techniques to incorporate external social norm knowledge bases can generally enhance performance. This work can facilitate the transition of robotics from user-guided passive manipulation to active social compliance.

Zheng Lian, Fan Zhang, Lan Chen et al.

Open-Vocabulary Multimodal Emotion Recognition (OV-MER) aims to predict emotions without being constrained by predefined label spaces, thereby enabling fine-grained emotion understanding. Unlike traditional discriminative methods, OV-MER leverages generative models to capture the full spectrum of emotions and employs emotion wheels (EWs) for metric calculation. Previous approaches primarily rely on token-level loss during training. However, this objective is misaligned with the metrics used in OV-MER, and these metrics cannot be directly optimized via gradient backpropagation. To address this limitation, we turn our attention to reinforcement learning, as this strategy can optimize non-differentiable objectives. We term this framework AffectGPT-RL. Furthermore, we conduct extensive experiments to elucidate the role of reinforcement learning in this task, revealing the necessity of the reasoning process, the impact of different rewards, and the generalizability to other emotion tasks such as sentiment analysis and basic emotion recognition. Experimental results demonstrate that AffectGPT-RL yields significant performance improvements on OV-MER. Beyond this task, we also achieve remarkable performance gains on basic emotion recognition, attaining state-of-the-art results on MER-UniBench. To the best of our knowledge, this is the pioneering work exploring the role of reinforcement learning in OV-MER, providing valuable guidance for subsequent researchers. Our code is provided in the supplementary material and will be released to facilitate future research.

Yuxin Zhang, Haoyu Chen, Zheng Lin et al.

Clustered federated learning (CFL) is proposed to mitigate the performance deterioration stemming from data heterogeneity in federated learning (FL) by grouping similar clients for cluster-wise model training. However, current CFL methods struggle due to inadequate integration of global and intra-cluster knowledge and the absence of an efficient online model similarity metric, while treating the cluster count as a fixed hyperparameter limits flexibility and robustness. In this paper, we propose an adaptive CFL framework, named FedAC, which (1) efficiently integrates global knowledge into intra-cluster learning by decoupling neural networks and utilizing distinct aggregation methods for each submodule, significantly enhancing performance; (2) includes a costeffective online model similarity metric based on dimensionality reduction; (3) incorporates a cluster number fine-tuning module for improved adaptability and scalability in complex, heterogeneous environments. Extensive experiments show that FedAC achieves superior empirical performance, increasing the test accuracy by around 1.82% and 12.67% on CIFAR-10 and CIFAR-100 datasets, respectively, under different non-IID settings compared to SOTA methods.

Chunjiang Liu, Xiaoyuan Wang, Qingran Lin et al.

We introduce the challenging problem of multi-object system identification from videos, for which prior methods are ill-suited due to their focus on single-object scenes or discrete material classification with a fixed set of material prototypes. To address this, we propose MOSIV, a new framework that directly optimizes for continuous, per-object material parameters using a differentiable simulator guided by geometric objectives derived from video. We also present a new synthetic benchmark with contact-rich, multi-object interactions to facilitate evaluation. On this benchmark, MOSIV substantially improves grounding accuracy and long-horizon simulation fidelity over adapted baselines, establishing it as a strong baseline for this new task. Our analysis shows that object-level fine-grained supervision and geometry-aligned objectives are critical for stable optimization in these complex, multi-object settings. The source code and dataset will be released.

Haoyu Chen, Wenbo Li, Jinjin Gu et al.

For image super-resolution (SR), bridging the gap between the performance on synthetic datasets and real-world degradation scenarios remains a challenge. This work introduces a novel "Low-Res Leads the Way" (LWay) training framework, merging Supervised Pre-training with Self-supervised Learning to enhance the adaptability of SR models to real-world images. Our approach utilizes a low-resolution (LR) reconstruction network to extract degradation embeddings from LR images, merging them with super-resolved outputs for LR reconstruction. Leveraging unseen LR images for self-supervised learning guides the model to adapt its modeling space to the target domain, facilitating fine-tuning of SR models without requiring paired high-resolution (HR) images. The integration of Discrete Wavelet Transform (DWT) further refines the focus on high-frequency details. Extensive evaluations show that our method significantly improves the generalization and detail restoration capabilities of SR models on unseen real-world datasets, outperforming existing methods. Our training regime is universally compatible, requiring no network architecture modifications, making it a practical solution for real-world SR applications.

Haoyu Chen, Xiaojie Xu, Wenbo Li et al.

Poster design is a critical medium for visual communication. Prior work has explored automatic poster design using deep learning techniques, but these approaches lack text accuracy, user customization, and aesthetic appeal, limiting their applicability in artistic domains such as movies and exhibitions, where both clear content delivery and visual impact are essential. To address these limitations, we present POSTA: a modular framework powered by diffusion models and multimodal large language models (MLLMs) for customized artistic poster generation. The framework consists of three modules. Background Diffusion creates a themed background based on user input. Design MLLM then generates layout and typography elements that align with and complement the background style. Finally, to enhance the poster's aesthetic appeal, ArtText Diffusion applies additional stylization to key text elements. The final result is a visually cohesive and appealing poster, with a fully modular process that allows for complete customization. To train our models, we develop the PosterArt dataset, comprising high-quality artistic posters annotated with layout, typography, and pixel-level stylized text segmentation. Our comprehensive experimental analysis demonstrates POSTA's exceptional controllability and design diversity, outperforming existing models in both text accuracy and aesthetic quality.

Yuxin Zhang, Haoyu Chen, Zheng Lin et al.

Clustered federated learning (CFL) addresses the performance challenges posed by data heterogeneity in federated learning (FL) by organizing edge devices with similar data distributions into clusters, enabling collaborative model training tailored to each group. However, existing CFL approaches strictly limit knowledge sharing to within clusters, lacking the integration of global knowledge with intra-cluster training, which leads to suboptimal performance. Moreover, traditional clustering methods incur significant computational overhead, especially as the number of edge devices increases. In this paper, we propose LCFed, an efficient CFL framework to combat these challenges. By leveraging model partitioning and adopting distinct aggregation strategies for each sub-model, LCFed effectively incorporates global knowledge into intra-cluster co-training, achieving optimal training performance. Additionally, LCFed customizes a computationally efficient model similarity measurement method based on low-rank models, enabling real-time cluster updates with minimal computational overhead. Extensive experiments show that LCFed outperforms state-of-the-art benchmarks in both test accuracy and clustering computational efficiency.

Yunlong Lin, Zixu Lin, Haoyu Chen et al.

Vision-centric perception systems struggle with unpredictable and coupled weather degradations in the wild. Current solutions are often limited, as they either depend on specific degradation priors or suffer from significant domain gaps. To enable robust and autonomous operation in real-world conditions, we propose JarvisIR, a VLM-powered agent that leverages the VLM as a controller to manage multiple expert restoration models. To further enhance system robustness, reduce hallucinations, and improve generalizability in real-world adverse weather, JarvisIR employs a novel two-stage framework consisting of supervised fine-tuning and human feedback alignment. Specifically, to address the lack of paired data in real-world scenarios, the human feedback alignment enables the VLM to be fine-tuned effectively on large-scale real-world data in an unsupervised manner. To support the training and evaluation of JarvisIR, we introduce CleanBench, a comprehensive dataset consisting of high-quality and large-scale instruction-responses pairs, including 150K synthetic entries and 80K real entries. Extensive experiments demonstrate that JarvisIR exhibits superior decision-making and restoration capabilities. Compared with existing methods, it achieves a 50% improvement in the average of all perception metrics on CleanBench-Real. Project page: https://cvpr2025-jarvisir.github.io/.

Yuanchang Liang, Xiaobo Wang, Kai Wang et al.

In Vision-Language-Action (VLA) models, action chunking (i.e., executing a sequence of actions without intermediate replanning) is a key technique to improve robotic manipulation abilities. However, a large chunk size reduces the model's responsiveness to new information, while a small one increases the likelihood of mode-jumping, jerky behavior resulting from discontinuities between chunks. Therefore, selecting the optimal chunk size is an urgent demand to balance the model's reactivity and consistency. Unfortunately, a dominant trend in current VLA models is an empirical fixed chunk length at inference-time, hindering their superiority and scalability across diverse manipulation tasks. To address this issue, we propose a novel Adaptive Action Chunking (AAC) strategy, which exploits action entropy as the cue to adaptively determine the chunk size based on current predictions. Extensive experiments on a wide range of simulated and real-world robotic manipulation tasks have demonstrated that our approach substantially improves performance over the state-of-the-art alternatives. The videos and source code are publicly available at https://lance-lot.github.io/adaptive-chunking.github.io/.