Concept-based models have been proposed as a new line of research for explainable by-design deep learning models. However, those models show their whole power when applied to benchmarks where the concepts are well defined and the concepts' attributes easily extractable from the raw data. In this paper, we challenge the most recent concept-based model initially developed for image classification, on more complex interpretative tasks from a recently proposed video benchmark where they perform poorly. We conduct a root cause analysis of the poor performances of state-of-the-art explainable concept-based models for these multimodal interpretative tasks, and propose adaptations to design robust explainable models for detecting character objectification in this novel challenging video benchmark. We show that the optimal architectural choice may vary depending on the modality setting, thereby showing that designing multimodal concept-based approaches remains an open challenge and calls for further investigation.
论文检索
输入标题、作者或关键词，从 7,537 篇学术成果中精准定位
The future of digital marketing lies in the convergence of human creativity and generative AI, where insight, strategy, and storytelling are co-authored by intelligent systems. We present MindFuse, a brave new explainable generative AI framework designed to act as a strategic partner in the marketing process. Unlike conventional LLM applications that stop at content generation, MindFuse fuses CTR-based content AI-guided co-creation with large language models to extract, interpret, and iterate on communication narratives grounded in real advertising data. MindFuse operates across the full marketing lifecycle: from distilling content pillars and customer personas from competitor campaigns to recommending in-flight optimizations based on live performance telemetry. It uses attention-based explainability to diagnose ad effectiveness and guide content iteration, while aligning messaging with strategic goals through dynamic narrative construction and storytelling. We introduce a new paradigm in GenAI for marketing, where LLMs not only generate content but reason through it, adapt campaigns in real time, and learn from audience engagement patterns. Our results, validated in agency deployments, demonstrate up to 12× efficiency gains, setting the stage for future integration with empirical audience data (e.g., GWI, Nielsen) and full-funnel attribution modeling. MindFuse redefines AI not just as a tool, but as a collaborative agent in the creative and strategic fabric of modern marketing. In the end of the paper, we also provide a forward-looking forecast on how platforms like MindFuse and Google DeepMind's ACAI are likely to shape the future of marketing agencies. These systems will not only reset client expectations toward greater transparency, speed, and personalization, but will also redefine the skillsets demanded from the new generation of marketers. As hybrid agencies emerge-blending creative storytelling with data science and AI engineering-the competitive landscape will increasingly hinge on talent. In this new environment, professionals will be expected to pair human imagination with technical fluency, while agencies will need to reinvent themselves as AI facilitators and curators of brand authenticity amidst the ongoing talent war led by platforms such as Meta.
Recent advances in vision-language models have enabled rich semantic understanding across modalities. However, these encoding methods lack the ability to interpret or reason about the moral dimensions of content---a crucial aspect of human cognition. In this paper, we address this gap by introducing MoralCLIP, a novel embedding representation method that extends multimodal learning with explicit moral grounding based on Moral Foundations Theory (MFT). Our approach integrates visual and textual moral cues into a unified embedding space, enabling cross-modal moral alignment. MoralCLIP is grounded on the multi-label dataset Social-Moral Image Database to identify co-occurring moral foundations in visual content. For MoralCLIP training, we design a moral data augmentation strategy to scale our annotated dataset to 15,000 image-text pairs labeled with MFT-aligned dimensions. Our results demonstrate that explicit moral supervision improves both unimodal and multimodal understanding of moral content, establishing a foundation for morally-aware AI systems capable of recognizing and aligning with human moral values.
Recent advancements in computer vision and Natural Language Processing (NLP) have transformed how machines interpret visual and textual information. While traditional image processing techniques rely on pixel-based analysis and feature extraction, and NLP leverages self-attention mechanisms for contextual understanding, the potential of viewing images as a language remains largely unexplored. In this work as a proof-of-concept, we introduce a novel ''Vision Language'' framework, transforming image pixels into sequences of alphanumeric characters, effectively treating images as ''text.'' This unique representation enables the direct application of NLP techniques to computer vision tasks, such as classification and segmentation, by utilizing text-based models for image analysis. We demonstrate this approach by adapting text classification methods for image classification on benchmark datasets. Furthermore, we construct a domain-specific medical language from the MedMNIST dataset, spanning 11 medical imaging sub-datasets, illustrating how a unified language can effectively generalize across distinct datasets. We create a broad, generalized language from ImageNet, leveraging approximately 1.3 million images to show that a language derived from extensive image data can simplify and unify image processing across tasks and datasets. Our findings reveal a promising cross-disciplinary approach that bridges computer vision and NLP. Our code is publicly available at https://github.com/mustakinalam/The-Birth-of-Vision-Language.
Managing personal health data is a challenge in today's fragmented and institution-centric healthcare ecosystem. Individuals often lack meaningful control over their medical records, which are scattered across incompatible systems and formats. This vision paper presents Health +, a user-centric, multimodal health data management system that empowers individuals (including those with limited technical expertise) to upload, query, and share their data across modalities (e.g., text, images, reports). Rather than aiming for institutional overhaul, Health + emphasizes individual agency by providing intuitive interfaces and intelligent recommendations for accessing and sharing data. At the system level, it tackles the complexity of storing, integrating, and securing heterogeneous health records, ensuring both efficiency and privacy. By unifying multimodal data and prioritizing patients, Health + enables a more connected, interpretable, and user-controlled health information ecosystem.
Concept Bottleneck Models (CBMs) enhance the interpretability of AI systems, particularly by bridging visual input with human-understandable concepts, effectively acting as a form of multimodal interpretability model. However, existing CBMs typically assume static datasets, which fundamentally limits their adaptability to real-world, continuously evolving multimodal data streams. To address this, we define a novel continual learning task for CBMs: simultaneously handling concept-incremental and class-incremental learning. This task requires models to continuously acquire new concepts (often representing cross-modal attributes) and classes while robustly preserving previously learned knowledge. To tackle this challenging problem, we propose CONceptual Continual Incremental Learning (CONCIL), a novel framework that fundamentally re-imagines concept and decision layer updates as linear regression problems. This reformulation eliminates the need for gradient-based optimization, thereby effectively preventing catastrophic forgetting. Crucially, CONCIL relies solely on recursive matrix operations, rendering it highly computationally efficient and well-suited for real-time and large-scale multimodal data applications. Experimental results compellingly demonstrate that CONCIL achieves ''absolute knowledge memory'' and significantly surpasses the performance of traditional CBM methods in both concept- and class-incremental settings, thus establishing a new paradigm for continual learning in CBMs, particularly valuable for dynamic multimodal understanding.
Commonaiverse is an interactive installation exploring human emotions through full-body motion tracking and real-time AI feedback. Participants engage in three phases: Teaching, Exploration and the Cosmos Phase, collaboratively expressing and interpreting emotions with the system. The installation integrates MoveNet for precise motion tracking and a multi-recommender AI system to analyze emotional states dynamically, responding with adaptive audiovisual outputs. By shifting from top-down emotion classification to participant-driven, culturally diverse definitions, we highlight new pathways for inclusive, ethical affective computing. We discuss how this collaborative, out-of-the-box approach pushes multimedia research beyond single-user facial analysis toward a more embodied, co-created paradigm of emotional AI. Furthermore, we reflect on how this reimagined framework fosters user agency, reduces bias, and opens avenues for advanced interactive applications.
Accurate accident anticipation remains challenging when driver cognition and dynamic road conditions are underrepresented in predictive models. In this paper, we propose CAMERA (Context-Aware Multi-modal Enhanced Risk Anticipation), a multi-modal framework integrating dashcam video, textual annotations, and driver attention maps for robust accident anticipation. Unlike existing methods that rely on static or environment-centric thresholds, CAMERA employs an adaptive mechanism guided by scene complexity and gaze entropy, reducing false alarms while maintaining high recall in dynamic, multi-agent traffic scenarios. A hierarchical fusion pipeline with Bi-GRU (Bidirectional GRU) captures spatio-temporal dependencies, while a Geo-Context Vision-Language module translates 3D spatial relationships into interpretable, human-centric alerts. Evaluations on the DADA-2000 and benchmarks show that CAMERA achieves state-of-the-art performance, improving accuracy and lead time. These results demonstrate the effectiveness of modeling driver attention, contextual description, and adaptive risk thresholds to enable more reliable accident anticipation.
Over the past decade, adaptive video streaming technology has witnessed significant advancements, particularly driven by the rapid evolution of deep learning techniques. However, the black-box nature of deep learning algorithms presents challenges for developers in understanding decision-making processes and optimizing for specific application scenarios. Although existing research has enhanced algorithm interpretability through decision tree conversion, interpretability does not directly equate to developers' subjective comprehensibility. To address this challenge, we introduce ComTree, the first bitrate adaptation algorithm generation framework that considers comprehensibility. The framework initially generates the complete set of decision trees that meet performance requirements, then leverages large language models to evaluate these trees for developer comprehensibility, ultimately selecting solutions that best facilitate human understanding and enhancement. Experimental results demonstrate that ComTree significantly improves comprehensibility while maintaining competitive performance, showing potential for further advancement. The source code and appendix are available at https://github.com/thu-media/ComTree.
The dynamic propagation of social media has broadened the reach of financial advisory content through podcast videos, yet extracting insights from lengthy, multimodal segments (30-40 minutes) remains challenging. We introduce FASTER(Financial Advisory Summariser with Textual Embedded Relevant images), a modular framework that tackles three key challenges: (1) extracting modality-specific features, (2) producing optimized, concise summaries, and (3) aligning visual keyframes with associated textual points. FASTER employs BLIP-2 for semantic visual descriptions, OCR for textual patterns, and Whisper-based transcription with Speaker diarization as BOS features. A modified Direct Preference Optimization (DPO)-based loss function, equipped with BOS-specific fact-checking, ensures precision, relevance, and factual consistency against the human-aligned summary. A ranker-based retrieval mechanism further aligns keyframes with summarized content, enhancing interpretability and cross-modal coherence. To acknowledge data resource scarcity, we introduce Fin-APT, a dataset comprising 470 publicly accessible financial advisory pep-talk videos for robust multimodal research. Comprehensive cross-domain experiments confirm FASTER's strong performance, robustness, and generalizability when compared to Large Language Models (LLMs) and Vision-Language Models (VLMs). By establishing a new standard for multimodal summarization, FASTER makes financial advisory content more accessible and actionable, thereby opening new avenues for research.
The rapid advancement of AI-generation models has enabled the creation of hyperrealistic imagery, posing ethical risks through widespread misinformation. Current deepfake detection methods, categorized as face-specific detectors or general AI-generated detectors, lack transparency by framing detection as a classification task without explaining decisions. While several LLM-based approaches offer explainability, they suffer from coarse-grained analyses and dependency on labor-intensive annotations. This paper introduces RAIDX (Retrieval-Augmented Image Deepfake Detection and Explainability), a novel deepfake detection framework integrating Retrieval-Augmented Generation (RAG) and Group Relative Policy Optimization (GRPO) to enhance detection accuracy and decision explainability. Specifically, RAIDX leverages RAG to incorporate external knowledge for improved detection accuracy and employs GRPO to autonomously generate fine-grained textual explanations and saliency maps, eliminating the need for extensive manual annotations. Experiments on multiple benchmarks demonstrate RAIDX's effectiveness in identifying real or fake, and providing interpretable rationales in both textual descriptions and saliency maps, achieving state-of-the-art detection performance while advancing transparency in deepfake identification. RAIDX represents the first unified framework to synergize RAG and GRPO, addressing critical gaps in accuracy and explainability. Our code and models will be publicly available.
From Prediction to Explanation: Multimodal, Explainable, and Interactive Deepfake Detection Framework for Non-Expert Users
PDF ↗The proliferation of deepfake technologies poses urgent challenges and serious risks to digital integrity, particularly within critical sectors such as forensics, journalism, and the legal system. While existing detection systems have made significant progress in classification accuracy, they typically function as black-box models, offering limited transparency and minimal support for human reasoning. This hinders their usability in real-world decision-making contexts, especially for non-expert users. We present DF-P2E (Deepfake: Prediction to Explanation), a novel multimodal framework that integrates visual, semantic, and narrative layers of explanation to make deepfake detection interpretable and accessible. The framework consists of three modular components: (1) a deepfake classifier with Grad-CAM-based saliency visualisation, (2) a visual captioning module that generates natural language summaries of manipulated regions, and (3) a narrative refinement module that uses a fine-tuned LLM to produce context-aware, user-sensitive explanations. We instantiate and evaluate the framework on the DF40 benchmark, the most diverse deepfake dataset to date. Experiments demonstrate that our system achieves competitive detection performance while providing high-quality explanations aligned with Grad-CAM activations. By unifying prediction and explanation in a coherent, human-aligned pipeline, this work offers a scalable approach to interpretable deepfake detection, advancing the broader vision of trustworthy and transparent AI systems for media forensics.
Text-to-Image (T2I) diffusion models exhibit concerning tendencies to generate harmful imagery that perpetuates social biases and stereotypes, posing significant ethical risks in real-world applications. While existing mitigation approaches predominantly employ black-box methodologies through dataset augmentation or constrained fine-tuning, they face critical limitations, including high data acquisition costs and potential exacerbation of stereotypes during model retraining. Inspired by neuroscience principles where neurological dysfunction often stems from aberrant neural activation patterns, we propose a novel framework, StereoClinic, targeting the root cause of stereotype generation through direct neural intervention. Our solution introduces two synergistic components: Diffusion Deep Taylor Decomposition (DDTD) for precisely localizing stereotype-related neurons via Layer-wise Relevance Propagation (LRP) attribution analysis, and Stereotype Neuron Suppression (SNS) implementing targeted activation damping to neutralize bias propagation. Through extensive empirical evaluations across multiple bias dimensions, we demonstrate that our method achieves significant stereotype mitigation without compromising image quality or requiring additional training data. This neuro-inspired approach establishes a new paradigm for model interpretability and ethical alignment in generative AI systems.
Assessing whether AI-generated images are substantially similar to copyrighted works is a crucial step in resolving copyright disputes. In this paper, we propose CopyJudge, an automated copyright infringement identification framework that leverages large vision-language models (LVLMs) to simulate practical court processes for determining substantial similarity between copyrighted images and those generated by text-to-image diffusion models. Specifically, we employ an abstraction-filtration-comparison test framework with multi-LVLM debate to assess the likelihood of infringement and provide detailed judgment rationales. Based on the judgments, we further introduce a general LVLM-based mitigation strategy that automatically optimizes infringing prompts by avoiding sensitive expressions while preserving the non-infringing content. Besides, our approach can be enhanced by exploring non-infringing noise vectors within the diffusion latent space via reinforcement learning, even without modifying the original prompts.Experimental results show that our identification method achieves comparable state-of-the-art performance, while offering superior generalization and interpretability across various forms of infringement, and that our mitigation method could more effectively mitigate memorization and IP infringement without losing non-infringing expressions.
With the rapid advancement of video generation models such as Sora, video quality assessment (VQA) is becoming increasingly crucial for selecting high-quality videos from large-scale datasets used in pre-training. Traditional VQA methods, typically producing single numerical scores, often lack comprehensiveness and interpretability. To address these challenges, we introduce MVQA-68K, a novel multi-dimensional VQA dataset comprising over 68,000 carefully annotated videos, covering seven essential quality dimensions: overall aesthetics, camera movement, dynamic degree, texture detail, composition, visual quality, and factual consistency. Each annotation includes detailed chain-of-thought reasoning to facilitate interpretability and comprehensive understanding. Extensive experiments demonstrate that MVQA-68K significantly enhances the performance of various multimodal large language models (MLLMs) on the VQA task, achieving state-of-the-art results not only on our internal test set (Fig.1) but also on public benchmarks including LSVQ-test, LSVQ-1080p, and LIVE-VQC. Meantime, incorporating explicit reasoning process during VQA training substantially boosts the zero-shot generalization. Code and dataset will be available at github: https://github.com/Controller01-ai/MVQA-68K.
Timely and accurate diagnosis of neurodegenerative disorders, such as Alzheimer's disease, is central to disease management. Existing deep learning models require large annotated datasets and often act as ''black boxes''. However, clinical datasets are frequently small or lack labels, limiting the effectiveness of these methods. Here, we introduce REMEMBER - Retrieval-based Explainable Multimodal Evidence-guided Modeling for Brain Evaluation and Reasoning - a machine learning framework that enables zero- and few-shot Alzheimer's diagnosis from brain MRI scans via reference-based reasoning. Specifically, REMEMBER first contrastively trains a vision-text model on expert-annotated reference data, using pseudo-text modalities to encode abnormality types, diagnosis labels, and composite clinical descriptions. At inference time, it retrieves similar, human-validated cases from a curated dataset and integrates their contextual information via an evidence encoder and attention-based inference head. This evidence-guided design allows REMEMBER to mimic clinical decision-making by grounding predictions in retrieved imaging and textual context. It outputs diagnostic predictions with an interpretable report, including reference images and clinical-aligned explanations. Experimental results demonstrate that REMEMBER achieves robust zero- and few-shot performance and offers a powerful and explainable framework to neuroimaging-based diagnosis in the real world, especially under limited data.
Multivariate time-series (MTS) classification tasks play a key role in data-driven applications spanning healthcare, finance, and mobile communication. As MTS data are typically collected from multiple interdependent sensors, the resulting temporal patterns inherently reflect the characteristics of the underlying sensing systems. Despite this connection, conventional MTS classification models predominantly focus on raw time-series data while disregarding valuable sensor-specific prior knowledge, which fundamentally constrains their classification accuracy. The emergence of large language models (LLMs) has encoded extensive sensor-related knowledge within their parameter spaces. However, effectively harnessing such knowledge to enhance MTS classification networks remains an open challenge. To address this, we propose Foresail, a status-guided neural framework that bridges this gap through systematic integration of LLM-derived sensor knowledge via the status relationship matrix and fine-grained status labels. Foresail can be seamlessly integrated with existing MTS networks to optimize performance and generate interpretable intermediate results. Experiments on irregularly and regularly sampled MTS data demonstrate that Foresail outperforms state-of-the-art approaches, achieving a notable improvement in F1-score of up to 10.9% compared to the basic MTS network.
Domain generalization (DG) plays a pivotal role in enabling models to maintain robust performance across heterogeneous environments. However, existing DG methods are fundamentally constrained by two intertwined limitations: (1) causal misalignment, which stems from undifferentiated feature encoding that entangles causal mechanisms with environmental biases; (2)semantic conflict arises when conventional adaptation methods find it challenging to balance the preservation of class discriminability with the mitigation of domain-specific distribution discrepancies. To address these challenges of DG, we propose a novel Causal-Driven Semantic Consistency Reasoning (CauRDG) method, which synergistically integrates Prototype-Guided Causal Disentanglement (PGCD) and Dual-Space Semantic Disambiguation (DSSD). Specifically, PGCD constructs a causal framework that identifies stable relationships and decouples invariant mechanisms from domain-specific variations, preserving causal consistency while adapting to contextual differences. DSSD harnesses a dual-space paradigm, enhancing local categorical clarity and maintaining global conceptual unity, thus balancing domain-specific precision with cross-domain coherence. The robustness provided by CauRDG ensures robust extraction and interpretation of essential features by preserving invariant causal structures, thereby harmonizing discriminative semantics with domain-varying contexts. Extensive experiments on multiple benchmark datasets consistently demonstrate the effectiveness and superiority of our CauRDG over state-of-the-art baselines.
The emergence of multimodal large language models has redefined the agent paradigm by integrating language and vision modalities with external data sources, enabling agents to better interpret human instructions and execute increasingly complex tasks. However, in this paper, we identify a critical yet previously overlooked security vulnerability in multimodal agents: cross-modal prompt injection attacks. To exploit this vulnerability, we propose CrossInject, a novel attack framework in which attacker embeds adversarial perturbations across multiple modalities to align with target malicious content, allowing external instructions to hijack the agents' decision-making process and execute unauthorized tasks. Our approach incorporates two key coordinated components. First, we introduce Visual Latent Alignment, where we optimize adversarial features to the malicious instructions in the visual embedding space based on a text-to-image generative model, ensuring that adversarial images subtly encode cues for malicious task execution. Subsequently, we present Textual Guidance Enhancement, where a large language model is leveraged to construct the black-box defensive system prompt through adversarial meta-prompting and generate a malicious textual command based on it that steers the agents' output toward better compliance with attacker's requests. Extensive experiments demonstrate that our method outperforms state-of-the-art attacks, achieving at least a +30.1% increase in attack success rates across diverse tasks. Furthermore, we validate our attack's effectiveness in real-world multimodal autonomous agents, highlighting its potential implications for safety-critical applications. Code can be found in https://github.com/Larry0454/CrossInject.
Medical Foundation Models (MFMs) are revolutionizing radiography image analysis with scalable and generalized diagnostic capabilities. However, their effectiveness in real-world clinical practice is limited due to insufficient interpretability. To address this limitation, we propose RadLAS, a novel MFM for interpretable Radiographic image analysis by introducing Lesion-Aware Self-supervised pre-training. Unlike conventional MFMs that rely on post-hoc explanations, RadLAS innovates by directly emulating human diagnostic reasoning to first grounding lesion evidence and then making decisions accordingly. Specifically, RadLAS introduces two self-supervised tasks: (I) Lesion-grounded Reconstruction, which learns structured anatomical representations by restoring lesion-aware image patches into their healthy counterparts, thereby facilitating pixel-level grounding of lesion evidence via input-normal contrast. (II) Lesion-discrimination Contrastive Learning, which enhances lesion-aware pattern in representations by explicitly decoupling grounded lesion evidence as clinical cues and aligning them with global semantics, thereby enabling direct lesion-oriented diagnosis while preserving global context. RadLAS demonstrates excellent performance across diverse downstream radiographic datasets, offering verifiable explanations by deriving specific diagnoses (Task II) based on grounded lesion evidence (Task I), while preserving generalized representations essential for high diagnostic accuracy. Extensive experiments demonstrate that RadLAS (i) achieves superior interpretability with highly correlated lesion prediction and localization, surpassing 11 interpretable medical models; (ii) delivers scalable representation learning, outperforming 14 SOTA supervised and self-supervised MFMs.