Yuan Gao

Chao Deng, Jiale Yuan, Pi Bu et al.

Large vision language models (LVLMs) have improved the document understanding capabilities remarkably, enabling the handling of complex document elements, longer contexts, and a wider range of tasks. However, existing document understanding benchmarks have been limited to handling only a small number of pages and fail to provide a comprehensive analysis of layout elements locating. In this paper, we first define three primary task categories: Long Document Understanding, numerical Reasoning, and cross-element Locating, and then propose a comprehensive benchmark, LongDocURL, integrating above three primary tasks and comprising 20 sub-tasks categorized based on different primary tasks and answer evidences. Furthermore, we develop a semi-automated construction pipeline and collect 2,325 high-quality question-answering pairs, covering more than 33,000 pages of documents, significantly outperforming existing benchmarks. Subsequently, we conduct comprehensive evaluation experiments on both open-source and closed-source models across 26 different configurations, revealing critical performance gaps in this field.

Chunjiang Ge, Sijie Cheng, Ziming Wang et al.

High-resolution Large Multimodal Models (LMMs) encounter the challenges of excessive visual tokens and quadratic visual complexity. Current high-resolution LMMs address the quadratic complexity while still generating excessive visual tokens. However, the redundancy in visual tokens is the key problem as it leads to more substantial compute. To mitigate this issue, we propose ConvLLaVA, which employs ConvNeXt, a hierarchical backbone, as the visual encoder of LMM to replace Vision Transformer (ViT). ConvLLaVA compresses high-resolution images into information-rich visual features, effectively preventing the generation of excessive visual tokens. To enhance the capabilities of ConvLLaVA, we propose two critical optimizations. Since the low-resolution pretrained ConvNeXt underperforms when directly applied on high resolution, we update it to bridge the gap. Moreover, since ConvNeXt's original compression ratio is inadequate for much higher resolution inputs, we train a successive stage to further compress the visual tokens, thereby reducing redundancy. These optimizations enable ConvLLaVA to support inputs of 1536x1536 resolution generating only 576 visual tokens, capable of handling images of arbitrary aspect ratios. Experimental results demonstrate that our method achieves competitive performance with state-of-the-art models on mainstream benchmarks. The ConvLLaVA model series are publicly available at https://github.com/alibaba/conv-llava.

Yuan Gao, Haoping Bai, Zequn Jie et al.

We propose to incorporate neural architecture search (NAS) into general-purpose multi-task learning (GP-MTL). Existing NAS methods typically define different search spaces according to different tasks. In order to adapt to different task combinations (i.e., task sets), we disentangle the GP-MTL networks into single-task backbones (optionally encode the task priors), and a hierarchical and layerwise features sharing/fusing scheme across them. This enables us to design a novel and general task-agnostic search space, which inserts cross-task edges (i.e., feature fusion connections) into fixed single-task network backbones. Moreover, we also propose a novel single-shot gradient-based search algorithm that closes the performance gap between the searched architectures and the final evaluation architecture. This is realized with a minimum entropy regularization on the architecture weights during the search phase, which makes the architecture weights converge to near-discrete values and therefore achieves a single model. As a result, our searched model can be directly used for evaluation without (re-)training from scratch. We perform extensive experiments using different single-task backbones on various task sets, demonstrating the promising performance obtained by exploiting the hierarchical and layerwise features, as well as the desirable generalizability to different i) task sets and ii) single-task backbones. The code of our paper is available at https://github.com/bhpfelix/MTLNAS.

Daoze Zhang, Chenghan Fu, Zhanheng Nie et al.

With the rapid advancement of e-commerce, exploring general representations rather than task-specific ones has attracted increasing research attention. For product understanding, although existing discriminative dual-flow architectures drive progress in this field, they inherently struggle to model the many-to-one alignment between multiple images and texts of products. Therefore, we argue that generative Multimodal Large Language Models (MLLMs) hold significant potential for improving product representation learning. Nevertheless, achieving this goal still remains non-trivial due to several key challenges: the lack of multimodal and aspect-aware modeling modules in typical LLMs; the common presence of background noise in product images; and the absence of a standard benchmark for evaluation. To address these issues, we propose the first generative MLLM-based model named MOON for product representation learning. Our method (1) employs a guided Mixture-of-Experts (MoE) module for targeted modeling of multimodal and aspect-specific product content; (2) effectively detects core semantic regions in product images to mitigate the distraction and interference caused by background noise; and (3) introduces the specialized negative sampling strategy to increase the difficulty and diversity of negative samples. In addition, we release a large-scale multimodal benchmark MBE for various product understanding tasks. Experimentally, our model demonstrates competitive zero-shot performance on both our benchmark and the public dataset, showcasing strong generalization across various downstream tasks, including cross-modal retrieval, product classification, and attribute prediction. Furthermore, the case study and visualization illustrate the effectiveness of MOON for product understanding.

Huanshuo Dong, Hong Wang, Hao Wu et al.

Neural operators have demonstrated considerable effectiveness in accelerating the solution of time-dependent partial differential equations (PDEs) by directly learning governing physical laws from data. However, for PDEs governed by conservation laws(e.g., conservation of mass, energy, or matter), existing neural operators fail to satisfy conservation properties, which leads to degraded model performance and limited generalizability. Moreover, we observe that distinct PDE problems generally require different optimal neural network architectures. This finding underscores the inherent limitations of specialized models in generalizing across diverse problem domains. To address these limitations, we propose Exterior-Embedded Conservation Framework (ECF), a universal conserving framework that can be integrated with various data-driven neural operators to enforce conservation laws strictly in predictions. The framework consists of two key components: a conservation quantity encoder that extracts conserved quantities from input data, and a conservation quantity decoder that adjusts the neural operator's predictions using these quantities to ensure strict conservation compliance in the final output. Since our architecture enforces conservation laws, we theoretically prove that it enhances model performance. To validate the performance of our method, we conduct experiments on multiple conservation-law-constrained PDE scenarios, including adiabatic systems, shallow water equations, and the Allen-Cahn problem. These baselines demonstrate that our method effectively improves model accuracy while strictly enforcing conservation laws in the predictions.

Tassilo Wald, Ibrahim Ethem Hamamci, Yuan Gao et al.

Vision-language pre-training, i.e., aligning images with paired text, is a powerful paradigm to create encoders that can be directly used for tasks such as classification and retrieval, and for downstream tasks such as segmentation and report generation. In the 3D medical image domain, these capabilities allow vision-language encoders (VLEs) to support radiologists by retrieving patients with similar abnormalities or predicting likelihoods of abnormality. While the methodology holds promise, data availability limits the capabilities of current 3D VLEs. In this paper, we alleviate the lack of data by injecting additional inductive biases: introducing a report generation objective and pairing vision-language pre-training with vision-only pre-training. This allows us to leverage both image-only and paired image-text 3D datasets, increasing the total amount of data to which our model is exposed. Through these additional inductive biases, paired with best practices of the 3D medical imaging domain, we develop the Comprehensive Language-image Pre-training (COLIPRI) encoder family. Our COLIPRI encoders achieve state-of-the-art performance in report generation, classification probing, and zero-shot classification, and remain competitive for semantic segmentation.

Huadong Wu, Yi Fu, Yunhao Li et al.

Facial makeup editing aims to realistically transfer makeup from a reference to a target face. Existing methods often produce low-quality results with coarse makeup details and struggle to preserve both identity and makeup fidelity, mainly due to the lack of structured paired data -- where source and result share identity, and reference and result share identical makeup. To address this, we introduce MakeupQuad, a large-scale, high-quality dataset with non-makeup faces, references, edited results, and textual makeup descriptions. Building on this, we propose EvoMakeup, a unified training framework that mitigates image degradation during multi-stage distillation, enabling iterative improvement of both data and model quality. Although trained solely on synthetic data, EvoMakeup generalizes well and outperforms prior methods on real-world benchmarks. It supports high-fidelity, controllable, multi-task makeup editing -- including full-face and partial reference-based editing, as well as text-driven makeup editing -- within a single model. Experimental results demonstrate that our method achieves superior makeup fidelity and identity preservation, effectively balancing both aspects. Code and dataset will be released upon acceptance.

Yuan Gao, Alan Yuille

Many objects, especially these made by humans, are symmetric, e.g. cars and aeroplanes. This paper addresses the estimation of 3D structures of symmetric objects from multiple images of the same object category, e.g. different cars, seen from various viewpoints. We assume that the deformation between different instances from the same object category is non-rigid and symmetric. In this paper, we extend two leading non-rigid structure from motion (SfM) algorithms to exploit symmetry constraints. We model the both methods as energy minimization, in which we also recover the missing observations caused by occlusions. In particularly, we show that by rotating the coordinate system, the energy can be decoupled into two independent terms, which still exploit symmetry, to apply matrix factorization separately on each of them for initialization. The results on the Pascal3D+ dataset show that our methods significantly improve performance over baseline methods.

Yuan Gao, Jiayi Ma, Alan L. Yuille

This paper addresses the problem of face recognition when there is only few, or even only a single, labeled examples of the face that we wish to recognize. Moreover, these examples are typically corrupted by nuisance variables, both linear (i.e., additive nuisance variables such as bad lighting, wearing of glasses) and non-linear (i.e., non-additive pixel-wise nuisance variables such as expression changes). The small number of labeled examples means that it is hard to remove these nuisance variables between the training and testing faces to obtain good recognition performance. To address the problem we propose a method called Semi-Supervised Sparse Representation based Classification (S$^3$RC). This is based on recent work on sparsity where faces are represented in terms of two dictionaries: a gallery dictionary consisting of one or more examples of each person, and a variation dictionary representing linear nuisance variables (e.g., different lighting conditions, different glasses). The main idea is that (i) we use the variation dictionary to characterize the linear nuisance variables via the sparsity framework, then (ii) prototype face images are estimated as a gallery dictionary via a Gaussian Mixture Model (GMM), with mixed labeled and unlabeled samples in a semi-supervised manner, to deal with the non-linear nuisance variations between labeled and unlabeled samples. We have done experiments with insufficient labeled samples, even when there is only a single labeled sample per person. Our results on the AR, Multi-PIE, CAS-PEAL, and LFW databases demonstrate that the proposed method is able to deliver significantly improved performance over existing methods.

Yuan Gao, Alan L. Yuille

Many man-made objects have intrinsic symmetries and Manhattan structure. By assuming an orthographic projection model, this paper addresses the estimation of 3D structures and camera projection using symmetry and/or Manhattan structure cues, which occur when the input is single- or multiple-image from the same category, e.g., multiple different cars. Specifically, analysis on the single image case implies that Manhattan alone is sufficient to recover the camera projection, and then the 3D structure can be reconstructed uniquely exploiting symmetry. However, Manhattan structure can be difficult to observe from a single image due to occlusion. To this end, we extend to the multiple-image case which can also exploit symmetry but does not require Manhattan axes. We propose a novel rigid structure from motion method, exploiting symmetry and using multiple images from the same category as input. Experimental results on the Pascal3D+ dataset show that our method significantly outperforms baseline methods.