Zero-Shot Anomaly Detection (ZSAD) aims to identify and localize anomalous regions in images of unseen object classes. While recent methods based on vision-language models like CLIP show promise, their performance is constrained by existing prompt engineering strategies. Current approaches, whether relying on single fixed, learnable, or dense dynamic prompts, suffer from a representational bottleneck and are prone to overfitting on auxiliary data, failing to generalize to the complexity and diversity of unseen anomalies. To overcome these limitations, we propose PromptMoE. Our core insight is that robust ZSAD requires a compositional approach to prompt learning. Instead of learning monolithic prompts, PromptMoE learns a pool of expert prompts, which serve as a basis set of composable semantic primitives, and a visually-guided Mixture-of-Experts (MoE) mechanism to dynamically combine them for each instance. Our framework materializes this concept through a Visually-Guided Mixture of Prompt (VGMoP) that employs an image-gated sparse MoE to aggregate diverse normal and abnormal expert state prompts, generating semantically rich textual representations with strong generalization. Extensive experiments across 15 datasets in industrial and medical domains demonstrate the effectiveness and state-of-the-art performance of PromptMoE.

In the quest for artificial general intelligence, Multi-modal Large Language Models (MLLMs) have emerged as a focal point in recent advancements. However, the predominant focus remains on developing their capabilities in static image understanding. The potential of MLLMs to process sequential visual data is still insufficiently explored, highlighting the lack of a comprehensive, high-quality assessment of their performance. In this paper, we introduce Video-MME, the first-ever full-spectrum, Multi-Modal Evaluation benchmark of MLLMs in Video analysis. Our work distinguishes from existing benchmarks through four key features: 1) Diversity in video types, spanning 6 primary visual domains with 30 subfields to ensure broad scenario generalizability; 2) Duration in temporal dimension, encompassing both short-, medium-, and long-term videos, ranging from 11 seconds to 1 hour, for robust contextual dynamics; 3) Breadth in data modalities, integrating multi-modal inputs besides video frames, including subtitles and audios, to unveil the all-round capabilities of MLLMs; 4) Quality in annotations, utilizing rigorous manual labeling by expert annotators to facilitate precise and reliable model assessment. With Video-MME, we extensively evaluate various state-of-the-art MLLMs, and reveal that Gemini 1.5 Pro is the best-performing commercial model, significantly outperforming the open-source models with an average accuracy of 75%, compared to 71.9% for GPT-4o. The results also demonstrate that Video-MME is a universal benchmark that applies to both image and video MLLMs. Further analysis indicates that subtitle and audio information could significantly enhance video understanding. Besides, a decline in MLLM performance is observed as video duration increases for all models. Our dataset along with these findings underscores the need for further improvements in handling longer sequences and multi-modal data, shedding light on future MLLM development. Project page: https://video-mme.github.io.

The swift progress of Multi-modal Large Models (MLLMs) has showcased their impressive ability to tackle tasks blending vision and language. Yet, most current models and benchmarks cater to scenarios with a narrow scope of visual and textual contexts. These models often fall short when faced with complex comprehension tasks, which involve navigating through a plethora of irrelevant and potentially misleading information in both text and image forms. To bridge this gap, we introduce a new, more demanding task known as Interleaved Image-Text Comprehension (IITC). This task challenges models to discern and disregard superfluous elements in both images and text to accurately answer questions and to follow intricate instructions to pinpoint the relevant image. In support of this task, we further craft a new VEGA dataset, tailored for the IITC task on scientific content, and devised a subtask, Image-Text Association (ITA), to refine image-text correlation skills. Our evaluation of four leading closed-source models, as well as various open-source models using VEGA, underscores the rigorous nature of IITC. Even the most advanced models, such as Gemini-1.5-pro and GPT4V, only achieved modest success. By employing a multi-task, multi-scale post-training strategy, we have set a robust baseline for MLLMs on the IITC task, attaining an $85.8\%$ accuracy rate in image association and a $0.508$ Rouge score. These results validate the effectiveness of our dataset in improving MLLMs capabilities for nuanced image-text comprehension.

Multimodal Large Language Model (MLLM) relies on the powerful LLM to perform multimodal tasks, showing amazing emergent abilities in recent studies, such as writing poems based on an image. However, it is difficult for these case studies to fully reflect the performance of MLLM, lacking a comprehensive evaluation. In this paper, we fill in this blank, presenting the first comprehensive MLLM Evaluation benchmark MME. It measures both perception and cognition abilities on a total of 14 subtasks. In order to avoid data leakage that may arise from direct use of public datasets for evaluation, the annotations of instruction-answer pairs are all manually designed. The concise instruction design allows us to fairly compare MLLMs, instead of struggling in prompt engineering. Besides, with such an instruction, we can also easily carry out quantitative statistics. A total of 30 advanced MLLMs are comprehensively evaluated on our MME, which not only suggests that existing MLLMs still have a large room for improvement, but also reveals the potential directions for the subsequent model optimization. The data are released at the project page: https://github.com/BradyFU/Awesome-Multimodal-Large-Language-Models/tree/Evaluation.

Multi-modal Large Language Models (MLLMs) excel at single-image tasks but struggle with multi-image understanding due to cross-modal misalignment, leading to hallucinations (context omission, conflation, and misinterpretation). Existing methods using Direct Preference Optimization (DPO) constrain optimization to a solitary image reference within the input sequence, neglecting holistic context modeling. To address this, we propose Context-to-Cue Direct Preference Optimization (CcDPO), a multi-level preference optimization framework that enhances per-image perception in multi-image settings by zooming into visual clues—from sequential context to local details. Our approach features two sequentially dependent components: (i) Context-Level Optimization: By introducing low-cost sequence preference pairs, we optimize the model to distinguish between complete and disrupted multi-image contexts, thereby correcting cognitive biases in MLLMs’ multi-image understanding. (ii) Needle-Level Optimization: By integrating region-specific visual prompts with multimodal preference supervision, we direct the model’s attention to critical visual details, effectively suppressing perceptual biases toward fine-grained visual information. To support scalable optimization, we also construct MultiScope-42k, an automatically generated multi-image dataset with hierarchical preference pairs. Experiments show that CcDPO significantly reduces hallucinations and yields consistent performance gains across general single- and multi-image tasks. Codes are available at https://github.com/LXDxmu/CcDPO.

With the growing requirement for natural human-computer interaction, speech-based systems receive increasing attention as speech is one of the most common forms of daily communication. However, the existing speech models still experience high latency when generating the first audio token during streaming, which poses a significant bottleneck for deployment. To address this issue, we propose VITA-Audio, an end-to-end large speech model with fast audio-text token generation. Specifically, we introduce a lightweight Multiple Cross-modal Token Prediction (MCTP) module that efficiently generates multiple audio tokens within a single model forward pass, which not only accelerates the inference but also significantly reduces the latency for generating the first audio in streaming scenarios. In addition, a four-stage progressive training strategy is explored to achieve model acceleration with minimal loss of speech quality. To our knowledge, VITA-Audio is the first multi-modal large language model capable of generating audio output during the first forward pass, enabling real-time conversational capabilities with minimal latency. VITA-Audio is fully reproducible and is trained on open-source data only. Experimental results demonstrate that our model achieves an inference speedup of 3~5x at the 7B parameter scale, but also significantly outperforms open-source models of similar model size on multiple benchmarks for automatic speech recognition (ASR), text-to-speech (TTS), and spoken question answering (SQA) tasks.

Recently, transformer-based methods have achieved state-of-the-art prediction quality on human pose estimation(HPE). Nonetheless, most of these top-performing transformer-based models are too computation-consuming and storage-demanding to deploy on edge computing platforms. Those transformer-based models that require fewer resources are prone to under-fitting due to their smaller scale and thus perform notably worse than their larger counterparts. Given this conundrum, we introduce SD-Pose, a new self-distillation method for improving the performance of small transformer-based models. To mitigate the problem of under-fitting, we design a transformer module named Multi-Cycled Transformer(MCT) based on multiple-cycled forwards to more fully exploit the potential of small model parameters. Further, in order to prevent the additional inference compute-consuming brought by MCT, we introduce a self-distillation scheme, extracting the knowledge from the MCT module to a naive forward model. Specifically, on the MSCOCO validation dataset, SDPose-T obtains 69.7% mAP with 4.4M parameters and 1.8 GFLOPs. Furthermore, SDPose-S-V2 obtains 73.5% mAP on the MSCOCO validation dataset with 6.2M parameters and 4.7 GFLOPs, achieving a new state-of-the-art among predominant tiny neural network methods.

Vision foundation models have been explored recently to build general-purpose vision systems. However, predomi-nant paradigms, driven by casting instance-level tasks as an object-word alignment, bring heavy cross-modality in-teraction, which is not effective in prompting object detection and visual grounding. Another line of work that fo-cuses on pixel-level tasks often encounters a large annotation gap of things and stuff, and suffers from mutual inter-ference between foreground-object and background-class segmentation. In stark contrast to the prevailing methods, we present APE, a universal visual perception model for aligning and prompting everything all at once in an image to perform diverse tasks, i.e., detection, segmentation, and grounding, as an instance-level sentence-object matching paradigm. Specifically, APE advances the convergence of detection and grounding by reformulating language-guided grounding as open-vocabulary detection, which efficiently scales up model prompting to thousands of category vocab-ularies and region descriptions while maintaining the ef-fectiveness of cross-modality fusion. To bridge the granu-larity gap of different pixel-level tasks, APE equalizes se-mantic and panoptic segmentation to proxy instance learning by considering any isolated regions as individual in-stances. APE aligns vision and language representation on broad data with natural and challenging characteristics all at once without task-specific fine-tuning. The extensive ex-periments on over 160 datasets demonstrate that, with only one-suit of weights, APE outperforms (or is on par with) the state-of-the-art models, proving that an effective yet univer-sal perception for anything aligning and prompting is in-deed feasible. Codes and trained models are released at https://github.com/shenyunhang/APE.

With the advent of large language models(LLMs) enhanced by the chain-of-thought(CoT) methodology, the visual reasoning problem is usually decomposed into manageable sub-tasks and tackled sequentially with various external tools. However, such a paradigm faces the challenge of the potential "determining hallucinations" in decision generation due to insufficient visual information and the limitation of low-level perception tools that fail to provide abstract summaries necessary for comprehensive reasoning. We argue that converging visual context acquisition and logical reasoning is pivotal for tackling visual reasoning tasks. This paper delves into the realm of multimodal CoT to solve intricate visual reasoning tasks with multimodal large language models(MLLMs) and their cognitive capability. To this end, we propose an innovative multimodal CoT framework, termed Cantor, characterized by a perception-decision architecture. Cantor first acts as a decision generator and integrates visual inputs to analyze the image and problem, ensuring a closer alignment with the actual context. Furthermore, Cantor leverages the advanced cognitive functions of MLLMs to perform as multifaceted experts for deriving higher-level information, enhancing the CoT generation process. Our extensive experiments demonstrate the efficacy of the proposed framework, showing significant improvements in multimodal CoT performance across two complex visual reasoning datasets, without necessitating fine-tuning or ground-truth rationales. Project Page: https://ggg0919.github.io/cantor/.

The integration of large language models into open-world detection frameworks significantly improves versatility in new environments. Prompt representations derived from these models help establish classification boundaries for both base and novel categories within open-world detectors. However, we are the first to discover that directly fine-tuning language models in detection systems results in redundant attention patterns and leads to suboptimal prompt representations. In order to fully leverage the capabilities of large language models and augment prompt encoding for detection, this study introduces a redundancy assessment metric to identify uniform attention patterns. Furthermore, in areas with high redundancy, we incorporate multimodal inplace prompt tuning (MIPT) to enrich the text prompt with visual clues. Experimental results validate the efficacy of our MIPT framework, achieving a notable increase across benchmarks, e.g. elevating GLIP-L from 22.6% to 25.0% on ODinW-35, and 9.0% improvement on LVIS.

We introduce MQ-Det, an efficient architecture and pre-training strategy design to utilize both textual description with open-set generalization and visual exemplars with rich description granularity as category queries, namely, Multi-modal Queried object Detection, for real-world detection with both open-vocabulary categories and various granularity. MQ-Det incorporates vision queries into existing well-established language-queried-only detectors. A plug-and-play gated class-scalable perceiver module upon the frozen detector is proposed to augment category text with class-wise visual information. To address the learning inertia problem brought by the frozen detector, a vision conditioned masked language prediction strategy is proposed. MQ-Det's simple yet effective architecture and training strategy design is compatible with most language-queried object detectors, thus yielding versatile applications. Experimental results demonstrate that multi-modal queries largely boost open-world detection. For instance, MQ-Det significantly improves the state-of-the-art open-set detector GLIP by +7.8% AP on the LVIS benchmark via multi-modal queries without any downstream finetuning, and averagely +6.3% AP on 13 few-shot downstream tasks, with merely additional 3% modulating time required by GLIP. Code is available at https://github.com/YifanXu74/MQ-Det.

Unsupervised domain adaptation (UDA) aims to address the domain-shift problem between a labeled source domain and an unlabeled target domain. Many efforts have been made to eliminate the mismatch between the distributions of training and testing data by learning domain-invariant representations. However, the learned representations are usually not task-oriented, i.e., being class-discriminative and domain-transferable simultaneously. This drawback limits the flexibility of UDA in complicated open-set tasks where no labels are shared between domains. In this paper, we break the concept of task-orientation into task-relevance and task-irrelevance, and propose a dynamic task-oriented disentangling network (DTDN) to learn disentangled representations in an end-to-end fashion for UDA. The dynamic disentangling network effectively disentangles data representations into two components: the task-relevant ones embedding critical information associated with the task across domains, and the task-irrelevant ones with the remaining non-transferable or disturbing information. These two components are regularized by a group of task-specific objective functions across domains. Such regularization explicitly encourages disentangling and avoids the use of generative models or decoders. Experiments in complicated, open-set scenarios (retrieval tasks) and empirical benchmarks (classification tasks) demonstrate that the proposed method captures rich disentangled information and achieves superior performance.

Domain generalization (DG) offers a preferable real-world setting for Person Re-Identification (Re-ID), which trains a model using multiple source domain datasets and expects it to perform well in an unseen target domain without any model updating. Unfortunately, most DG approaches are designed explicitly for classification tasks, which fundamentally differs from the retrieval task Re-ID. Moreover, existing applications of DG in Re-ID cannot correctly handle the massive variation among Re-ID datasets. In this paper, we identify two fundamental challenges in DG for Person Re-ID: domain-wise variations and identity-wise similarities. To this end, we propose an end-to-end Dual Distribution Alignment Network (DDAN) to learn domain-invariant features with dual-level constraints: the domain-wise adversarial feature learning and the identity-wise similarity enhancement. These constraints effectively reduce the domain-shift among multiple source domains further while agreeing to real-world scenarios. We evaluate our method in a large-scale DG Re-ID benchmark and compare it with various cutting-edge DG approaches. Quantitative results show that DDAN achieves state-of-the-art performance.

Person Re-Identification (ReID) has achieved remarkable performance along with the deep learning era. However, most approaches carry out ReID only based upon holistic pedestrian regions. In contrast, real-world scenarios involve occluded pedestrians, which provide partial visual appearances and destroy the ReID accuracy. A common strategy is to locate visible body parts by auxiliary model, which however suffers from significant domain gaps and data bias issues. To avoid such problematic models in occluded person ReID, we propose the Occlusion-Aware Mask Network (OAMN). In particular, we incorporate an attention-guided mask module, which requires guidance from labeled occlusion data. To this end, we propose a novel occlusion augmentation scheme that produces diverse and precisely labeled occlusion for any holistic dataset. The proposed scheme suits real-world scenarios better than existing schemes, which consider only limited types of occlusions. We also offer a novel occlusion unification scheme to tackle ambiguity information at the test phase. The above three components enable existing attention mechanisms to precisely capture body parts regardless of the occlusion. Comprehensive experiments on a variety of person ReID benchmarks demonstrate the superiority of OAMN over state-of-the-arts.

The decision-based black-box attack means to craft adversarial examples with only the top-1 label of the victim model available. A common practice is to start from a large perturbation and then iteratively reduce it with a deterministic direction and a random one while keeping it adversarial. The limited information obtained from each query and inefficient direction sampling impede attack efficiency, making it hard to obtain a small enough perturbation within a limited number of queries. To tackle this problem, we propose a novel attack method termed Adaptive History-driven Attack (AHA) which gathers information from all historical queries as the prior for current sampling. Moreover, to balance between the deterministic direction and the random one, we dynamically adjust the coefficient according to the ratio of the actual magnitude reduction to the expected one. Such a strategy improves the success rate of queries during optimization, letting adversarial examples move swiftly along the decision boundary. Our method can also integrate with subspace optimization like dimension reduction to further improve efficiency. Extensive experiments on both ImageNet and CelebA datasets demonstrate that our method achieves at least 24.3% lower magnitude of perturbation on average with the same number of queries. Finally, we prove the practical potential of our method by evaluating it on popular defense methods and a real-world system provided by MEGVII Face++.

We are interested in exploring the possibility and benefits of structure learning for deep models. As the first step, this paper investigates the matter for Restricted Boltzmann Machines (RBMs). We conduct the study with Replicated Softmax, a variant of RBMs for unsupervised text analysis. We present a method for learning what we call Sparse Boltzmann Machines, where each hidden unit is connected to a subset of the visible units instead of all of them. Empirical results show that the method yields models with significantly improved model fit and interpretability as compared with RBMs where each hidden unit is connected to all visible units.

Hierarchical latent tree analysis (HLTA) is recently proposed as a new method for topic detection. It differs fundamentally from the LDA-based methods in terms of topic definition, topic-document relationship, and learning method. It has been shown to discover significantly more coherent topics and better topic hierarchies. However, HLTA relies on the Expectation-Maximization (EM) algorithm for parameter estimation and hence is not efficient enough to deal with large datasets. In this paper, we propose a method to drastically speed up HLTA using a technique inspired by the advances in the method of moments. Empirical experiments show that our method greatly improves the efficiency of HLTA. It is as efficient as the state-of-the-art LDA-based method for hierarchical topic detection and finds substantially better topics and topic hierarchies.

Low-rank matrix factorization has long been recognized as a fundamental problem in many computer vision applications. Nevertheless, the reliability of existing matrix factorization methods is often hard to guarantee due to challenges brought by such model selection issues as selecting the noise model and determining the model capacity. We address these two issues simultaneously in this paper by proposing a robust non-parametric Bayesian adaptive matrix factorization (AMF) model. AMF proposes a new noise model built on the Dirichlet process Gaussian mixture model (DP-GMM) by taking advantage of its high flexibility on component number selection and capability of fitting a wide range of unknown noise. AMF also imposes an automatic relevance determination (ARD) prior on the low-rank factor matrices so that the rank can be determined automatically without the need for enforcing any hard constraint. An efficient variational method is then devised for model inference. We compare AMF with state-of-the-art matrix factorization methods based on data sets ranging from synthetic data to real-world application data. From the results, AMF consistently achieves better or comparable performance.