In this work, we introduce Mini-Gemini, a simple and effective framework enhancing multi-modality Vision Language Models (VLMs). Despite the advancements in VLMs facilitating basic visual dialog and reasoning, a performance gap persists compared to advanced models like GPT-4 and Gemini. We propose a novel approach to narrow the gap by mining the potential of VLMs for better performance across various cross-modal tasks. It tackles the following questions: (1) How can high-resolution visual tokens improve image understanding without lengthening the token sequence? (2) How to improve reasoning and generation abilities of VLM with high-quality data? (3) How to close the gap between open-source VLMs and proprietary models on reasoning-driven generation? In particular, to enhance visual tokens, we propose to utilize an additional visual encoder for high-resolution refinement without increasing the visual token count. We further construct a high-quality dataset that promotes precise image comprehension and reasoning-based generation, expanding the operational scope of current VLMs. In general, Mini-Gemini further mines the potential of VLMs and empowers current frameworks with image understanding, reasoning, and generation simultaneously. The proposed model supports a series of dense and MoE Large Language Models (LLMs) from 2B to 34B, which achieve leading performance in several zero-shot benchmarks and even surpasses the developed private models. It is demonstrated to attain 80.6% accuracy on the MMB benchmark (+5.4 vs Gemini Pro) and 74.1% on TextVQA (+4.6 vs LLaVA-NeXT), achieving leading performance in several zero-shot benchmarks and even surpasses the developed private models. Furthermore, Mini-Gemini is proven to improve consistently with stronger LLM, visual encoder, and data in experiments.

As Multi-modal Large Language Models (MLLMs) evolve, expanding beyond single-domain capabilities is essential to meet the demands for more versatile and efficient AI. However, previous omni-models have insufficiently explored speech, neglecting its integration with multi-modality. We introduce Lyra, an efficient MLLM that enhances multimodal abilities, including advanced long-speech comprehension, sound understanding, cross-modality efficiency, and seamless speech interaction. To achieve efficiency and speech-centric capabilities, Lyra employs three strategies: (1) leveraging existing open-source large models and a proposed multi-modality LoRA to reduce training costs and data requirements; (2) using a latent multi-modality regularizer and extractor to strengthen the relationship between speech and other modalities, thereby enhancing model performance; and (3) constructing a high-quality, extensive dataset that includes 1.5M multi-modal (language, vision, audio) data samples and 12K long speech samples, enabling Lyra to handle complex long speech inputs and achieve more robust omni-cognition. Compared to other omni-methods, Lyra achieves state-of-the-art performance on various vision-language, vision-speech, and speech-language benchmarks, while also using fewer computational resources and less training data.

In this paper, we delve deeper into the Kullback–Leibler (KL) Divergence loss and mathematically prove that it is equivalent to the Decoupled Kullback-Leibler (DKL) Divergence loss that consists of 1) a weighted Mean Square Error ($\mathbf{w}$MSE) loss and 2) a Cross-Entropy loss incorporating soft labels. Thanks to the decomposed formulation of DKL loss, we have identified two areas for improvement. Firstly, we address the limitation of KL/DKL in scenarios like knowledge distillation by breaking its asymmetric optimization property. This modification ensures that the $\mathbf{w}$MSE component is always effective during training, providing extra constructive cues. Secondly, we introduce class-wise global information into KL/DKL to mitigate bias from individual samples. With these two enhancements, we derive the Improved Kullback–Leibler (IKL) Divergence loss and evaluate its effectiveness by conducting experiments on CIFAR-10/100 and ImageNet datasets, focusing on adversarial training, and knowledge distillation tasks. The proposed approach achieves new state-of-the-art adversarial robustness on the public leaderboard --- \textit{RobustBench} and competitive performance on knowledge distillation, demonstrating the substantial practical merits. Our code is available at https://github.com/jiequancui/DKL.

In this paper, we propose the Generalized Parametric Contrastive Learning (GPaCo/PaCo) which works well on both imbalanced and balanced data. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnable centers to rebalance from an optimization perspective. Further, we analyze our GPaCo/PaCo loss under a balanced setting. Our analysis demonstrates that GPaCo/PaCo can adaptively enhance the intensity of pushing samples of the same class close as more samples are pulled together with their corresponding centers and benefit hard example learning. Experiments on long-tailed benchmarks manifest the new state-of-the-art for long-tailed recognition. On full ImageNet, models from CNNs to vision transformers trained with GPaCo loss show better generalization performance and stronger robustness compared with MAE models. Moreover, GPaCo can be applied to semantic segmentation task and obvious improvements are observed on 4 most popular benchmarks.

A recent study has shown a phenomenon called neural collapse in that the within-class means of features and the classifier weight vectors converge to the vertices of a simplex equiangular tight frame at the terminal phase of training for classification. In this paper, we explore the cor-responding structures of the last-layer feature centers and classifiers in semantic segmentation. Based on our empirical and theoretical analysis, we point out that semantic segmentation naturally brings contextual correlation and imbalanced distribution among classes, which breaks the equiangular and maximally separated structure of neural collapse for both feature centers and classifiers. However, such a symmetric structure is beneficial to discrimination for the minor classes. To preserve these advantages, we in-troduce a regularizer on feature centers to encourage the network to learn features closer to the appealing structure in imbalanced semantic segmentation. Experimental results show that our method can bring significant improvements on both 2D and 3D semantic segmentation bench-marks. Moreover, our method ranks 1stand sets a new record (+6.8% mIoU) on the ScanNet200 test leaderboard.

Deep learning algorithms face great challenges with long-tailed data distribution which, however, is quite a common case in real-world scenarios. Previous methods tackle the problem from either the aspect of input space (re-sampling classes with different frequencies) or loss space (re-weighting classes with different weights), suffering from heavy over-fitting to tail classes or hard optimization during training. To alleviate these issues, we propose a more fundamental perspective for long-tailed recognition, i.e., from the aspect of parameter space, and aims to preserve specific capacity for classes with low frequencies. From this perspective, the trivial solution utilizes different branches for the head, medium, tail classes respectively, and then sums their outputs as the final results is not feasible. Instead, we design the effective residual fusion mechanism - with one main branch optimized to recognize images from all classes, another two residual branches are gradually fused and optimized to enhance images from medium+tail classes and tail classes respectively. Then the branches are aggregated into final results by additive shortcuts. We test our method on several benchmarks, i.e., long-tailed version of CIFAR-10, CIFAR-100, Places, ImageNet, and iNaturalist 2018. Experimental results manifest the effectiveness of our method. Our code is available at https://github.com/jiequancui/ResLT.

Deep neural networks may perform poorly when training datasets are heavily class-imbalanced. Recently, two-stage methods decouple representation learning and classifier learning to improve performance. But there is still the vital issue of miscalibration. To address it, we design two methods to improve calibration and performance in such scenarios. Motivated by the fact that predicted probability distributions of classes are highly related to the numbers of class instances, we propose label-aware smoothing to deal with different degrees of over-confidence for classes and improve classifier learning. For dataset bias between these two stages due to different samplers, we further propose shifted batch normalization in the decoupling framework. Our proposed methods set new records on multiple popular long-tailed recognition benchmark datasets, including CIFAR-10-LT, CIFAR-100-LT, ImageNet-LT, Places-LT, and iNaturalist 2018.

In this paper, we propose Parametric Contrastive Learning (PaCo) to tackle long-tailed recognition. Based on theoretical analysis, we observe supervised contrastive loss tends to bias on high-frequency classes and thus increases the difficulty of imbalanced learning. We introduce a set of parametric class-wise learnable centers to rebalance from an optimization perspective. Further, we analyze our PaCo loss under a balanced setting. Our analysis demonstrates that PaCo can adaptively enhance the intensity of pushing samples of the same class close as more samples are pulled together with their corresponding centers and benefit hard example learning. Experiments on long-tailed CIFAR, ImageNet, Places, and iNaturalist 2018 manifest the new state-of-the-art for long-tailed recognition. On full ImageNet, models trained with PaCo loss surpass supervised contrastive learning across various ResNet backbones, e.g., our ResNet-200 achieves 81.8% top-1 accuracy. Our code is available at https://github.com/dvlab-research/Parametric-Contrastive-Learning.

Deep image compression systems mainly contain four components: encoder, quantizer, entropy model, and decoder. To optimize these four components, a joint rate-distortion framework was proposed, and many deep neural network-based methods achieved great success in image compression. However, almost all convolutional neural network-based methods treat channel-wise feature maps equally, reducing the flexibility in handling different types of information. In this paper, we propose a channel-level variable quantization network to dynamically allocate more bitrates for significant channels and withdraw bitrates for negligible channels. Specifically, we propose a variable quantization controller. It consists of two key components: the channel importance module, which can dynamically learn the importance of channels during training, and the splitting-merging module, which can allocate different bitrates for different channels. We also formulate the quantizer into a Gaussian mixture model manner. Quantitative and qualitative experiments verify the effectiveness of the proposed model and demonstrate that our method achieves superior performance and can produce much better visual reconstructions.

Self-attention mechanism has been widely used for various tasks. It is designed to compute the representation of each position by a weighted sum of the features at all positions. Thus, it can capture long-range relations for computer vision tasks. However, it is computationally consuming. Since the attention maps are computed w.r.t all other positions. In this paper, we formulate the attention mechanism into an expectation-maximization manner and iteratively estimate a much more compact set of bases upon which the attention maps are computed. By a weighted summation upon these bases, the resulting representation is low-rank and deprecates noisy information from the input. The proposed Expectation-Maximization Attention (EMA) module is robust to the variance of input and is also friendly in memory and computation. Moreover, we set up the bases maintenance and normalization methods to stabilize its training procedure. We conduct extensive experiments on popular semantic segmentation benchmarks including PASCAL VOC, PASCAL Context, and COCO Stuff, on which we set new records1.

Cross-modal hashing encodes the multimedia data into a common binary hash space in which the correlations among the samples from different modalities can be effectively measured. Deep cross-modal hashing further improves the retrieval performance as the deep neural networks can generate more semantic relevant features and hash codes. In this paper, we study the unsupervised deep cross-modal hash coding and propose Deep Joint-Semantics Reconstructing Hashing (DJSRH), which has the following two main advantages. First, to learn binary codes that preserve the neighborhood structure of the original data, DJSRH constructs a novel joint-semantics affinity matrix which elaborately integrates the original neighborhood information from different modalities and accordingly is capable to capture the latent intrinsic semantic affinity for the input multi-modal instances. Second, DJSRH later trains the networks to generate binary codes that maximally reconstruct above joint-semantics relations via the proposed reconstructing framework, which is more competent for the batch-wise training as it reconstructs the specific similarity value unlike the common Laplacian constraint merely preserving the similarity order. Extensive experiments demonstrate the significant improvement by DJSRH in various cross-modal retrieval tasks.

Recently, a number of learning-based optimization methods that combine data-driven architectures with the classical optimization algorithms have been proposed and explored, showing superior empirical performance in solving various ill-posed inverse problems, but there is still a scarcity of rigorous analysis about the convergence behaviors of learning-based optimization. In particular, most existing analyses are specific to unconstrained problems but cannot apply to the more general cases where some variables of interest are subject to certain constraints. In this paper, we propose Differentiable Linearized ADMM (D-LADMM) for solving the problems with linear constraints. Specifically, D-LADMM is a K-layer LADMM inspired deep neural network, which is obtained by firstly introducing some learnable weights in the classical Linearized ADMM algorithm and then generalizing the proximal operator to some learnable activation function. Notably, we rigorously prove that there exist a set of learnable parameters for D-LADMM to generate globally converged solutions, and we show that those desired parameters can be attained by training D-LADMM in a proper way. To the best of our knowledge, we are the first to provide the convergence analysis for the learning-based optimization method on constrained problems.

Improving information flow in deep networks helps to ease the training difficulties and utilize parameters more efficiently. Here we propose a new convolutional neural network architecture with alternately updated clique (CliqueNet). In contrast to prior networks, there are both forward and backward connections between any two layers in the same block. The layers are constructed as a loop and are updated alternately. The CliqueNet has some unique properties. For each layer, it is both the input and output of any other layer in the same block, so that the information flow among layers is maximized. During propagation, the newly updated layers are concatenated to re-update previously updated layer, and parameters are reused for multiple times. This recurrent feedback structure is able to bring higher level visual information back to refine low-level filters and achieve spatial attention. We analyze the features generated at different stages and observe that using refined features leads to a better result. We adopt a multiscale feature strategy that effectively avoids the progressive growth of parameters. Experiments on image recognition datasets including CIFAR-10, CIFAR-100, SVHN and ImageNet show that our proposed models achieve the state-of-the-art performance with fewer parameters.

Convolutional neural networks (CNNs) have recently achieved great success in single-image super-resolution (SISR). However, these methods tend to produce over-smoothed outputs and miss some textural details. To solve these problems, we propose the Super-Resolution CliqueNet (SRCliqueNet) to reconstruct the high resolution (HR) image with better textural details in the wavelet domain. The proposed SRCliqueNet firstly extracts a set of feature maps from the low resolution (LR) image by the clique blocks group. Then we send the set of feature maps to the clique up-sampling module to reconstruct the HR image. The clique up-sampling module consists of four sub-nets which predict the high resolution wavelet coefficients of four sub-bands. Since we consider the edge feature properties of four sub-bands, the four sub-nets are connected to the others so that they can learn the coefficients of four sub-bands jointly. Finally we apply inverse discrete wavelet transform (IDWT) to the output of four sub-nets at the end of the clique up-sampling module to increase the resolution and reconstruct the HR image. Extensive quantitative and qualitative experiments on benchmark datasets show that our method achieves superior performance over the state-of-the-art methods.