Foundation models are pre-trained on large-scale datasets and subsequently fine-tuned on small-scale datasets using parameter-efficient fine-tuning (PEFT) techniques like low-rank adapters (LoRA). In most previous works, LoRA weight matrices are randomly initialized with a fixed rank across all attachment points. In this paper, we improve convergence and final performance of LoRA fine-tuning, using our proposed data-driven weight initialization method, ConsNoTrainLoRA (CNTLoRA). We express LoRA initialization as a domain shift problem where we use multiple constraints relating the pre-training and fine-tuning activations. By reformulating these constraints, we obtain a closed-form estimate of LoRA weights that depends on pre-training weights and fine-tuning activation vectors and hence requires no training during initialization. This weight estimate is decomposed to initialize the up and down matrices with proposed flexibility of variable ranks. With the proposed initialization method, we fine-tune on downstream tasks such as image generation, image classification and image understanding. Both quantitative and qualitative results demonstrate that CNTLoRA outperforms standard and data-driven weight initialization methods. Extensive analyses and ablations further elucidate the design choices of our framework, providing an optimal recipe for faster convergence and enhanced performance
论文检索
输入标题、作者或关键词，从 1,144 篇学术成果中精准定位
Despite significant advancements in image segmentation and object detection, understanding complex scenes remains a significant challenge. Here, we focus on graphical humor as a paradigmatic example of image interpretation that requires elucidating the interaction of different scene elements in the context of prior cognitive knowledge. This paper introduces HumorDB, a novel, controlled, and carefully curated dataset designed to evaluate and advance visual humor understanding by AI systems. The dataset comprises diverse images spanning photos, cartoons, sketches, and AI-generated content, including minimally contrastive pairs where subtle edits differentiate between humorous and non-humorous versions. We evaluate humans, state-of-the-art vision models, and large vision-language models on three tasks: binary humor classification, funniness rating prediction, and pairwise humor comparison. The results reveal a gap between current AI systems and human-level humor understanding. While pretrained vision-language models perform better than vision-only models, they still struggle with abstract sketches and subtle humor cues. Analysis of attention maps shows that even when models correctly classify humorous images, they often fail to focus on the precise regions that make the image funny. Preliminary mechanistic interpretability studies and evaluation of model explanations provide initial insights into how different architectures process humor. Our results identify promising trends and current limitations, suggesting that an effective understanding of visual humor requires sophisticated architectures capable of detecting subtle contextual features and bridging the gap between visual perception and abstract reasoning.All the code and data are available here: https://anonymous.4open.science/r/HumorDB_-049A
Humans recognize objects after observing only a few examples, a remarkable capability enabled by their inherent language understanding of the real-world environment. Developing verbalized and interpretable representation can significantly improve model generalization in low-data settings. In this work, we propose Verbalized Representation Learning (VRL), a novel approach for automatically extracting human-interpretable features for object recognition using few-shot data. Our method uniquely captures inter-class differences and intra-class commonalities in the form of natural language by employing a Vision-Language Model (VLM) to identify key discriminative features between different classes and shared characteristics within the same class. These verbalized features are then mapped to numeric vectors through the VLM. The resulting feature vectors can be further utilized to train and infer with downstream classifiers. Experimental results show that, at the same model scale, VRL achieves a 24% absolute improvement over prior state-of-the-art methods while using 95% less data and a smaller model. Furthermore, compared to human-labeled attributes, the features learned by VRL exhibit a 20% absolute gain when used for downstream classification tasks.
Revisiting Efficient Semantic Segmentation: Learning Offsets for Better Spatial and Class Feature Alignment
PDF ↗Semantic segmentation is fundamental to vision systems requiring pixel-level scene understanding, yet deploying it on resource-constrained devices demands efficient architectures. Although existing methods achieve real-time inference through lightweight designs, we reveal their inherent limitation: misalignment between class representations and image features caused by a per-pixel classification paradigm. With experimental analysis, we find that this paradigm results in a highly challenging assumption for efficient scenarios: Image pixel features should not vary for the same category in different images. To address this dilemma, we propose a coupled dual-branch offset learning paradigm that explicitly learns feature and class offsets to dynamically refine both class representations and spatial image features. Based on the proposed paradigm, we construct an efficient semantic segmentation network, OffSeg. Notably, the offset learning paradigm can be adopted to existing methods with no additional architectural changes. Extensive experiments on four datasets, including ADE20K, Cityscapes, COCO-Stuff-164K, and Pascal Context, demonstrate consistent improvements with negligible parameters.
Designing effective foundation models requires high-quality evaluation datasets. With the emergence of audio-visual foundation models, reliable assessment of their multi-modal understanding is essential. The current gold standard for evaluating audio-visual understanding is the popular classification dataset VGGSound. However, our analysis identifies several critical issues in VGGSound, including incomplete labelling, partially overlapping classes, and misaligned modalities. These flaws lead to distorted evaluations of models' true auditory and visual capabilities. To address these limitations, we introduce VGGSounder, a comprehensively re-annotated, multi-label test set extending VGGSound that is explicitly designed to accurately evaluate audio-visual foundation models. VGGSounder features detailed modality annotations, enabling precise analyses of modality-specific performance and revealing previously unnoticed model limitations. We believe VGGSounder offers a robust and reliable benchmark supporting the future development of audio-visual foundation models.
From Objects to Events: Unlocking Complex Visual Understanding in Object Detectors via LLM-guided Symbolic Reasoning
PDF ↗Current object detectors excel at entity localization and classification, yet exhibit inherent limitations in event recognition capabilities. This deficiency arises from their architecture's emphasis on discrete object identification rather than modeling the compositional reasoning, inter-object correlations, and contextual semantics essential for comprehensive event understanding. To address this challenge, we present a novel framework that expands the capability of standard object detectors beyond mere object recognition to complex event understanding through LLM-guided symbolic reasoning. Our key innovation lies in bridging the semantic gap between object detection and event understanding without requiring expensive task-specific training. The proposed plug-and-play framework interfaces with any open-vocabulary detector while extending their inherent capabilities across architectures. At its core, our approach combines (i) a symbolic regression mechanism exploring relationship patterns among detected entities and (ii) a LLM-guided strategically guiding the search toward meaningful expressions. These discovered symbolic rules transform low-level visual perception into interpretable event understanding, providing a transparent reasoning path from objects to events with strong transferability across domains.We compared our training-free framework against specialized event recognition systems across diverse application domains. Experiments demonstrate that our framework enhances multiple object detector architectures to recognize complex events such as illegal fishing activities (75 %AUROC, +8.36 %improvement), construction safety violations (+15.77%), and abnormal crowd behaviors (+23.16%).
Understanding 3D cell shape is crucial in biomedical research, where morphology serves as a key indicator of disease, cellular state, and drug response. However, many existing 3D point cloud classification models lack interpretability, limiting their utility for extracting biologically meaningful insights. In this work, we unify standard point cloud backbones and feature aggregation strategies within a Multiple Instance Learning (MIL) framework to enable inherently interpretable classification. Our approach, PointMIL, improves classification performance while providing fine-grained point-level explanations without relying on post hoc analysis. We demonstrate state-of-the-art mACC (97.3%) and F1 (97.5%) in the IntrA biomedical dataset and evaluate the interpretability using quantitative and qualitative metrics. Additionally, we introduce ATLAS-1, a novel dataset of drug-treated 3D cancer cells, and use it to show how PointMIL captures fine-grained morphological effects of chemical treatments. Beyond biomedical applications, PointMIL generalises to standard benchmarks such as ModelNet40 and ScanObjectNN, offering interpretable 3D object recognition across domains
HQ-CLIP: Leveraging Large Vision-Language Models to Create High-Quality Image-Text Datasets and CLIP Models
PDF ↗Large-scale but noisy image-text pair data have paved the way for the success of Contrastive Language-Image Pretraining (CLIP). As the foundation vision encoder, CLIP in turn serves as the cornerstone for most large vision-language models (LVLMs). This interdependence naturally raises an interesting question: Can we reciprocally leverage LVLMs to enhance the quality of image-text pair data, thereby opening the possibility of a self-reinforcing cycle for continuous improvement? In this work, we take a significant step toward this vision by introducing an LVLM-driven data refinement pipeline. Our framework leverages LVLMs to process images and their raw alt-text, generating four complementary textual formulas: long positive descriptions, long negative descriptions, short positive tags, and short negative tags. Applying this pipeline to the curated DFN-Large dataset yields VLM-150M, a refined dataset enriched with multi-grained annotations. Based on this dataset, we further propose a training paradigm that extends conventional contrastive learning by incorporating negative descriptions and short tags as additional supervised signals. The resulting model, namely HQ-CLIP, demonstrates remarkable improvements across diverse benchmarks. Within a comparable training data scale, our approach achieves state-of-the-art performance in zero-shot classification, cross-modal retrieval, and fine-grained visual understanding tasks. In retrieval benchmarks, HQ-CLIP even surpasses standard CLIP models trained on the DFN-2B dataset, which contains 10xmore training data than ours. All code, data, and models will be made publicly available to support further research.
3D semantic occupancy and flow prediction are fundamental to spatiotemporal scene understanding. This paper proposes a vision-based framework with three targeted improvements. First, we introduce an occlusion-aware adaptive lifting mechanism incorporating depth denoising. This enhances the robustness of 2D-to-3D feature transformation while mitigating reliance on depth priors. Second, we enforce 3D-2D semantic consistency via jointly optimized prototypes, using confidence- and category-aware sampling to address the long-tail classes problem. Third, to streamline joint prediction, we devise a BEV-centric cost volume to explicitly correlate semantic and flow features, supervised by a hybrid classification-regression scheme that handles diverse motion scales. Our purely convolutional architecture establishes new SOTA performance on multiple benchmarks for both semantic occupancy and joint occupancy semantic-flow prediction. We also present a family of models offering a spectrum of efficiency-performance trade-offs. Our real-time version exceeds all existing real-time methods in speed and accuracy, ensuring its practical viability.
Machine learning for image classification is an active and rapidly developing field. With the proliferation of classifiers of different sizes and different architectures, the problem of choosing the right model becomes more and more important. While we can assess a model's classification accuracy statistically, our understanding of the way these models work is unfortunately quite limited. In order to gain insight into the decision-making process of different vision models, we propose using minimal sufficient pixels sets. These pixels capture the essence of an image through the lens of the model. By comparing position, overlap and size of sets of pixels, we identify that different architectures have statistically different minimal pixels sets, in both size and position. In particular, ConvNext and EVA models differ markedly from the others. We also identify that images which are misclassified are associated with statistically significant larger pixels sets than correct classifications.
Although large visual-language models (LVLMs) have demonstrated strong performance in multimodal tasks, errors may occasionally arise due to biases during the reasoning process. Recently, reward models (RMs) have become increasingly pivotal in the reasoning process. Specifically, process RMs evaluate each reasoning step, outcome RMs focus on the assessment of reasoning results, and critique RMs perform error analysis on the entire reasoning process, followed by corrections. However, existing benchmarks for vision-language RMs (VLRMs) typically assess only a single aspect of their capabilities (e.g., distinguishing between two answers), thus limiting the all-round evaluation and restricting the development of RMs in the visual-language domain. To address this gap, we propose a comprehensive and challenging benchmark, dubbed as VLRMBench, encompassing 12,634 questions. VLRMBench is constructed based on three distinct types of datasets, covering mathematical reasoning, hallucination understanding, and multi-image understanding. We design 12 tasks across three major categories, focusing on evaluating VLRMs in the aspects of process understanding, outcome judgment, and critique generation. Extensive experiments are conducted on 21 open-source models and 5 advanced closed-source models, highlighting the challenges posed by VLRMBench. For instance, in the `Forecasting Future', a binary classification task, the advanced GPT-4o achieves only a 76.0% accuracy. The code is available at https://github.com/JCruan519/VLRMBench.
OrderChain: Towards General Instruct-Tuning for Stimulating the Ordinal Understanding Ability of MLLM
PDF ↗Despite the remarkable progress of multimodal large language models (MLLMs), they continue to face challenges in achieving competitive performance on ordinal regression (OR; a.k.a. ordinal classification). To address this issue, this paper presents OrderChain, a novel and general prompting paradigm that improves the ordinal understanding ability of MLLMs by specificity and commonality modeling. Specifically, our OrderChain consists of a set of task-aware prompts to facilitate the specificity modeling of diverse OR tasks and a new range optimization Chain-of-Thought (RO-CoT), which learns a commonality way of thinking about OR tasks by uniformly decomposing them into multiple small-range optimization subtasks. Further, we propose a category recursive division (CRD) method to generate instruction candidate category prompts to support RO-CoT automatic optimization. Comprehensive experiments show that LLaVA model with our OrderChain improves baseline LLaVA significantly on diverse OR datasets, e.g., from 47.5% to 93.2% accuracy on the Adience dataset for age estimation, and from 30.0% to 85.7% accuracy on the Diabetic Retinopathy dataset. Notably, LLaVA with our OrderChain also remarkably outperforms state-of-the-art methods by 27% on accuracy and 0.24 on MAE on the Adience dataset. To our best knowledge, our OrderChain is the first work that augments MLLMs for OR tasks, and the effectiveness is witnessed across a spectrum of OR datasets. Project Page: https://order-chain.github.io/.
Efficient lightweight neural networks have received increasing attention due to their faster reasoning speed and easier deployment on mobile devices. However, existing video models still focus on the larger ViT architecture, and few works attempt to build efficient architecture. Since many efficient contrastive language-image pre-training (CLIP) models have shown strong zero-shot classification and retrieval capability, we attempt to fill the gap in video-text understanding models and propose a fast and efficient video-text model MobileViCLIP with strong zero-shot reasoning capability that can be deployed on mobile devices. In particular, our MobileViCLIP-Small obtains similar zero-shot retrieval performance as InternVideo2-L14 on text-to-video dataset MSR-VTT while being 46.7x faster when deployed on the mobile device. Furthermore, MobileViCLIP-Small can generalize to zero-shot action recognition task and obtains 1.0% better Top-1 accuracy than InternVideo2-S14 while being 5.6x faster on the mobile device. The code is available at https://github.com/MCG-NJU/MobileViCLIP.
Adaptive Dual Uncertainty Optimization: Boosting Monocular 3D Object Detection under Test-Time Shifts
PDF ↗Accurate monocular 3D object detection (M3OD) is pivotal for safety-critical applications like autonomous driving, yet its reliability deteriorates significantly under real-world domain shifts caused by environmental or sensor variations. To address these shifts, Test-Time Adaptation (TTA) methods have emerged, enabling models to adapt to target distributions during inference. While prior TTA approaches recognize the positive correlation between low uncertainty and high generalization ability, they fail to address the dual uncertainty inherent to M3OD: semantic uncertainty (ambiguous class predictions) and geometric uncertainty (unstable spatial localization). To bridge this gap, we propose Dual Uncertainty Optimization (DUO), the first TTA framework designed to jointly minimize both uncertainties for robust M3OD. Through a convex optimization lens, we introduce an innovative convex structure of the focal loss and further derive a novel conjugate loss, enabling label-agnostic uncertainty weighting and balanced learning for high-uncertainty objects. In parallel, we design a semantic-aware normal field constraint that preserves geometric coherence in regions with clear semantic cues, reducing uncertainty from the unstable 3D representation. This dual-branch mechanism forms a complementary loop: enhanced spatial perception improves semantic classification, and robust semantic predictions further refine spatial understanding. Extensive experiments demonstrate the superiority of DUO over existing methods across various datasets and domain shift types. The source code is available at https://github.com/hzcar/DUO.
The deployment of deep learning models in critical domains necessitates a balance between high accuracy and interpretability.We introduce SIC, an inherently interpretable neural network that provides local and global explanations of its decision-making process.Leveraging the concept of case-based reasoning, SIC extracts class-representative support vectors from training images, ensuring they capture relevant features while suppressing irrelevant ones.Classification decisions are made by calculating and aggregating similarity scores between these support vectors and the input's latent feature vector. We employ B-Cos transformations, which align model weights with inputs, to yield coherent pixel-level explanations in addition to global explanations of case-based reasoning.We evaluate SIC on three tasks: fine-grained classification on Stanford Dogs and FunnyBirds, multi-label classification on Pascal VOC, and pathology detection on the RSNA dataset.Results indicate that SIC not only achieves competitive accuracy compared to state-of-the-art black-box and inherently interpretable models but also offers insightful explanations verified through practical evaluation on the FunnyBirds benchmark.Our theoretical analysis proves that these explanations fulfill established axioms for explanations. Our findings underscore SIC's potential for applications where understanding model decisions is as critical as the decisions themselves.
What Changed and What Could Have Changed? State-Change Counterfactuals for Procedure-Aware Video Representation Learning
PDF ↗Understanding a procedural activity requires modeling both how action steps transform the scene, and how evolving scene transformations can influence the sequence of action steps, even those that are accidental or erroneous. Existing work has studied procedure-aware video representations by modeling the temporal order of actions, but has not explicitly learned the state changes (scene transformations). In this work, we study procedure-aware video representation learning by incorporating state-change descriptions generated by Large Language Models (LLMs) as supervision signals for video encoders. Moreover, we generate state-change counterfactuals that simulate hypothesized failure outcomes, allowing models to learn by imagining unseen "What if" scenarios. This counterfactual reasoning facilitates the model's ability to understand the cause and effect of each step in an activity. We conduct extensive experiments on procedure-aware tasks, including temporal action segmentation, error detection, action phase classification, frame retrieval, multi-instance retrieval, and action recognition. Our results demonstrate the effectiveness of the proposed state-change descriptions and their counterfactuals, and achieve significant improvements on multiple tasks. Code is available at https://github.com/HCIS-Lab/counterfactual-video-pretrain.
Diffusion models have revolutionized generative modeling, enabling unprecedented realism in image and video synthesis.This success has sparked interest in leveraging their representations for visual understanding tasks. While recent works have explored this potential for image generation, the visual understanding capabilities of video diffusion models remain largely uncharted. To address this gap, we analyze the performance of latent image and video diffusion representations on various downstream tasks including image classification, action recognition, depth estimation, and tracking. For the most informative comparison, we utilize the same model architecture, WALT, across image and video generation objectives. Our results show that video generation pre-training consistently outperforms its image counterpart, though we find a striking range in the extent of this superiority. We further analyze features extracted from different layers and with varying noise levels, as well as the effect of model size and training budget on representation and generation quality. This work marks the first direct comparison of video and image diffusion objectives for visual understanding, offering insights into the role of temporal information in representation learning.
Deep Neural Networks (DNNs) have significantly advanced the field of computer vision. To improve DNN training process, knowledge distillation methods demonstrate their effectiveness in accelerating network training by introducing a fixed learning direction from the teacher network to student networks. In this context, several distillation-based optimization strategies are proposed, e.g., deep mutual learning and self-distillation, as an attempt to achieve generic training performance enhancement through the cooperative training of multiple networks. However, such strategies achieve limited improvements due to the poor understanding of the impact of learning directions among networks across different iterations. In this paper, we propose a novel competitive distillation strategy that allows each network in a group to potentially act as a teacher based on its performance, enhancing the overall learning performance. Competitive distillation organizes a group of networks to perform a shared task and engage in competition, where competitive optimization is proposed to improve the parameter updating process. We further introduce stochastic perturbation in competitive distillation, aiming to motivate networks to induce mutations to achieve better visual representations and global optimum. The experimental results show that competitive distillation achieves promising performance in diverse tasks and datasets.
Constructing Ophthalmic MLLM for Positioning-diagnosis Collaboration Through Clinical Cognitive Chain Reasoning
PDF ↗Multimodal large language models (MLLMs) demonstrate significant potential in the field of medical diagnosis. However, they face critical challenges in specialized domains such as ophthalmology, particularly the fragmentation of annotation granularity and inconsistencies in clinical reasoning logic, which hinder precise cross-modal understanding. This paper introduces FundusExpert, an ophthalmology-specific MLLM with integrated positioning-diagnosis reasoning capabilities, along with FundusGen, a dataset constructed through the intelligent Fundus-Engine system. Fundus-Engine automates localization and leverages MLLM-based semantic expansion to integrate global disease classification, local object detection, and fine-grained feature analysis within a single fundus image. Additionally, by constructing a clinically aligned cognitive chain, it guides the model to generate interpretable reasoning paths. FundusExpert, fine-tuned with instruction data from FundusGen, achieves the best performance in ophthalmic question-answering tasks, surpassing the average accuracy of the 40B MedRegA by 26.6%. It also excels in zero-shot report generation tasks, achieving a clinical consistency of 77.0%, significantly outperforming GPT-4o's 47.6%. Furthermore, we reveal a scaling law between data quality and model capability (L \propto N^ 0.068 ), demonstrating that the cognitive alignment annotations in FundusGen enhance data utilization efficiency. By integrating region-level localization with diagnostic reasoning chains, our work develops a scalable, clinically-aligned MLLM and explores a pathway toward bridging the visual-language gap in specific MLLMs. Our project can be found at https://github.com/MeteorElf/FundusExpert.
ConsistentCity: Semantic Flow-guided Occupancy DiT for Temporally Consistent Driving Scene Synthesis
PDF ↗Scene synthesis plays a crucial role in autonomous driving by addressing data scarcity and close-loop validation. Current approaches struggle to maintain temporal consistency in synthesized videos while preserving fine-grained details. We introduce ConsistentCity, a two-stage framework with a novel Semantic Flow-guided Diffusion Transformers (SF-DiT) that convert sequential BEV semantic maps into temporally consistent driving videos. Operating in a pretrained occupancy VQ-VAE latent space, our SF-DiT generates temporally consistent 3D occupancy, which provides guidance for controlled image and video diffusion for scene synthesis. To address the temporal consistency, SF-DiT enhances standard DiT blocks with temporal semantic modeling through two designs: (1) A Semantic Flow Estimation module capturing scene motions (flow, uncertainty, and classification) from sequential BEV semantic maps, and (2) A Semantic Flow-Modulated Cross-Attention module that dynamically adapts attention based on semantic flow patterns. This integration of semantic flow modeling in DiT enables consistent scene evolution understanding. Evaluations of image and video synthesis on nuScenes dataset demonstrate state-of-the-art performance with FID 8.3 and FVD 73.6, and superior temporal occupancy generation results on nuCraft and OpenOccupancy benchmarks.