With the rapid advancements in Artificial Intelligence Generated Image (AGI) technology, the accurate assessment of their quality has become an increasingly vital requirement. Prevailing methods typically rely on cross-modal models like CLIP or BLIP to evaluate text-image alignment and visual quality. However, when applied to AGIs, these methods encounter two primary challenges: semantic misalignment and details perception missing. To address these limitations, we propose Text-Visual Semantic Constrained AI-Generated Image Quality Assessment (SC-AGIQA), a unified framework that leverages text-visual semantic constraints to significantly enhance the comprehensive evaluation of both text-image consistency and perceptual distortion in AI-generated images. Our approach integrates key capabilities from multiple models and tackles the aforementioned challenges by introducing two core modules: the Text-assisted Semantic Alignment Module (TSAM), which leverages Multimodal Large Language Models (MLLMs) to bridge the semantic gap by generating an image description and comparing it against the original prompt for a refined consistency check, and the Frequency-domain Fine-Grained Degradation Perception Module (FFDPM), which draws inspiration from Human Visual System (HVS) properties by employing frequency domain analysis combined with perceptual sensitivity weighting to better quantify subtle visual distortions and enhance the capture of fine-grained visual quality details in images. Extensive experiments conducted on multiple benchmark datasets demonstrate that SC-AGIQA outperforms existing state-of-the-art methods. The code is publicly available at https://github.com/mozhu1/SC-AGIQA.
Despite their remarkable progress in multimodal understanding tasks, large vision language models (LVLMs) often suffer from "hallucinations", generating texts misaligned with the visual context. Existing methods aimed at reducing hallucinations through inference time intervention incur a significant increase in latency. To mitigate this, we present SPIN, a task-agnostic attention-guided head suppression strategy that can be seamlessly integrated during inference, without incurring any significant compute or latency overhead. We investigate whether hallucination in LVLMs can be linked to specific model components. Our analysis suggests that hallucinations can be attributed to a dynamic subset of attention heads in each layer. Leveraging this insight, for each text query token, we selectively suppress attention heads that exhibit low attention to image tokens, keeping the top-K attention heads intact. Extensive evaluations on visual question answering and image description tasks demonstrate the efficacy of SPIN in reducing hallucination scores up to 2.7x while maintaining F1, and improving throughput by 1.8x compared to existing alternatives. Code is available at https://github.com/YUECHE77/SPIN.
Recent advances in vision language models (VLMs) have enabled broad progress in the general medical field. However, pathology still remains a more challenging subdomain, with current pathology specific VLMs exhibiting limitations in both diagnostic accuracy and reasoning plausibility. Such shortcomings are largely attributable to the nature of current pathology datasets, which are primarily composed of image description pairs that lack the depth and structured diagnostic paradigms employed by real world pathologists. In this study, we leverage pathology textbooks and real world pathology experts to construct high-quality, reasoning-oriented datasets. Building on this, we introduce Patho-R1, a multimodal RL-based pathology Reasoner, trained through a three-stage pipeline: (1) continued pretraining on 3.5 million image-text pairs for knowledge infusion; (2) supervised fine-tuning on 500k high-quality Chain-of-Thought samples for reasoning incentivizing; (3) reinforcement learning using Group Relative Policy Optimization and Decoupled Clip and Dynamic sAmpling Policy Optimization strategies for multimodal reasoning quality refinement. To further assess the alignment quality of our dataset, we propose PathoCLIP, trained on the same figure-caption corpus used for continued pretraining. Comprehensive experimental results demonstrate that both PathoCLIP and Patho-R1 achieve robust performance across a wide range of pathology-related tasks, including zero-shot classification, cross-modal retrieval, Visual Question Answering, and Multiple Choice Question. Our project is available at the Patho-R1 repository: https://github.com/Wenchuan-Zhang/Patho-R1.
The examination of chest X-ray images is a crucial component in detecting various thoracic illnesses. This study introduces a new image description generation model that integrates a Vision Transformer (ViT) encoder with cross-modal attention and a GPT-4-based transformer decoder. The ViT captures high-quality visual features from chest X-rays, which are fused with text data through cross-modal attention to improve the accuracy, context, and richness of image descriptions. The GPT-4 decoder transforms these fused features into accurate and relevant captions. The model was tested on the National Institutes of Health (NIH) and Indiana University (IU) Chest X-ray datasets. On the IU dataset, it achieved scores of 0.854 (B-1), 0.883 (CIDEr), 0.759 (METEOR), and 0.712 (ROUGE-L). On the NIH dataset, it achieved the best performance on all metrics: BLEU 1--4 (0.825, 0.788, 0.765, 0.752), CIDEr (0.857), METEOR (0.726), and ROUGE-L (0.705). This framework has the potential to enhance chest X-ray evaluation, assisting radiologists in more precise and efficient diagnosis.
Rapid advancement of diffusion models has catalyzed remarkable progress in the field of image generation. However, prevalent models such as Flux, SD3.5 and Midjourney, still grapple with issues like model bias, limited text rendering capabilities, and insufficient understanding of Chinese cultural nuances. To address these limitations, we present Seedream 2.0, a native Chinese-English bilingual image generation foundation model that excels across diverse dimensions, which adeptly manages text prompt in both Chinese and English, supporting bilingual image generation and text rendering. We develop a powerful data system that facilitates knowledge integration, and a caption system that balances the accuracy and richness for image description. Particularly, Seedream is integrated with a self-developed bilingual large language model as a text encoder, allowing it to learn native knowledge directly from massive data. This enable it to generate high-fidelity images with accurate cultural nuances and aesthetic expressions described in either Chinese or English. Beside, Glyph-Aligned ByT5 is applied for flexible character-level text rendering, while a Scaled ROPE generalizes well to untrained resolutions. Multi-phase post-training optimizations, including SFT and RLHF iterations, further improve the overall capability. Through extensive experimentation, we demonstrate that Seedream 2.0 achieves state-of-the-art performance across multiple aspects, including prompt-following, aesthetics, text rendering, and structural correctness. Furthermore, Seedream 2.0 has been optimized through multiple RLHF iterations to closely align its output with human preferences, as revealed by its outstanding ELO score. In addition, it can be readily adapted to an instruction-based image editing model, such as SeedEdit, with strong editing capability that balances instruction-following and image consistency.
Deep Reinforcement Learning (DRL) has demonstrated strong performance in robotic control but remains susceptible to out-of-distribution (OOD) states, often resulting in unreliable actions and task failure. While previous methods have focused on minimizing or preventing OOD occurrences, they largely neglect recovery once an agent encounters such states. Although the latest research has attempted to address this by guiding agents back to in-distribution states, their reliance on uncertainty estimation hinders scalability in complex environments. To overcome this limitation, we introduce Language Models for Out-of-Distribution Recovery (LaMOuR), which enables recovery learning without relying on uncertainty estimation. LaMOuR generates dense reward codes that guide the agent back to a state where it can successfully perform its original task, leveraging the capabilities of LVLMs in image description, logical reasoning, and code generation. Experimental results show that LaMOuR substantially enhances recovery efficiency across diverse locomotion tasks and even generalizes effectively to complex environments, including humanoid locomotion and mobile manipulation, where existing methods struggle. The code and supplementary materials are available at https://lamour-rl.github.io/.
This research investigates both explicit and implicit social biases exhibited by Vision-Language Models (VLMs). The key distinction between these bias types lies in the level of awareness: explicit bias refers to conscious, intentional biases, while implicit bias operates subconsciously. To analyze explicit bias, we directly pose questions to VLMs related to gender and racial differences: (1) Multiple-choice questions based on a given image (e.g., "What is the education level of the person in the image?") (2) Yes-No comparisons using two images (e.g., "Is the person in the first image more educated than the person in the second image?") For implicit bias, we design tasks where VLMs assist users but reveal biases through their responses: (1) Image description tasks: Models are asked to describe individuals in images, and we analyze disparities in textual cues across demographic groups. (2) Form completion tasks: Models draft a personal information collection form with 20 attributes, and we examine correlations among selected attributes for potential biases. We evaluate Gemini-1.5, GPT-4V, GPT-4o, LLaMA-3.2-Vision and LLaVA-v1.6. Our code and data are publicly available at https://github.com/uscnlp-lime/VisBias.
Advancing AI in computational pathology requires large, high-quality, and diverse datasets, yet existing public datasets are often limited in organ diversity, class coverage, or annotation quality. To bridge this gap, we introduce SPIDER (Supervised Pathology Image-DEscription Repository), the largest publicly available patch-level dataset covering multiple organ types, including Skin, Colorectal, Thorax, and Breast with comprehensive class coverage for each organ. SPIDER provides high-quality annotations verified by expert pathologists and includes surrounding context patches, which enhance classification performance by providing spatial context. Alongside the dataset, we present baseline models trained on SPIDER using the Hibou-L foundation model as a feature extractor combined with an attention-based classification head. The models achieve state-of-the-art performance across multiple tissue categories and serve as strong benchmarks for future digital pathology research. Beyond patch classification, the model enables rapid identification of significant areas, quantitative tissue metrics, and establishes a foundation for multimodal approaches. Both the dataset and trained models are publicly available to advance research, reproducibility, and AI-driven pathology development. Access them at: https://github.com/HistAI/SPIDER
There is a growing need for diverse, high-quality stuttered speech data, particularly in the context of Indian languages. This paper introduces Project Boli, a multi-lingual stuttered speech dataset designed to advance scientific understanding and technology development for individuals who stutter, particularly in India. The dataset constitutes (a) anonymized metadata (gender, age, country, mother tongue) and responses to a questionnaire about how stuttering affects their daily lives, (b) captures both read speech (using the Rainbow Passage) and spontaneous speech (through image description tasks) for each participant and (c) includes detailed annotations of five stutter types: blocks, prolongations, interjections, sound repetitions and word repetitions. We present a comprehensive analysis of the dataset, including the data collection procedure, experience summarization of people who stutter, severity assessment of stuttering events and technical validation of the collected data. The dataset is released as an open access to further speech technology development.
Considering the lack of a unified framework for image description and deep cultural analysis at the subject level in the field of Ancient Chinese Paintings (ACP), this study utilized the Beijing Palace Museum's ACP collections to develop a semantic model integrating the iconological theory with a new workflow for term extraction and mapping. Our findings underscore the model's effectiveness. SDM can be used to support further art-related knowledge organization and cultural exploration of ACPs.
CLIP (Contrastive Language-Image Pre-training) has attained great success in pattern recognition and computer vision. Transferring CLIP to downstream tasks (e.g. zero- or few-shot classification) is a hot topic in multimodal learning. However, current studies primarily focus on either prompt learning for text or adapter tuning for vision, without fully exploiting the complementary information and correlations among image-text pairs. In this paper, we propose an Image Description Enhanced CLIP-Adapter (IDEA) method to adapt CLIP to few-shot image classification tasks. This method captures fine-grained features by leveraging both visual features and textual descriptions of images. IDEA is a training-free method for CLIP, and it can be comparable to or even exceeds state-of-the-art models on multiple tasks. Furthermore, we introduce Trainable-IDEA (T-IDEA), which extends IDEA by adding two lightweight learnable components (i.e., a projector and a learnable latent space), further enhancing the model's performance and achieving SOTA results on 11 datasets. As one important contribution, we employ the Llama model and design a comprehensive pipeline to generate textual descriptions for images of 11 datasets, resulting in a total of 1,637,795 image-text pairs, named "IMD-11". Our code and data are released at https://github.com/FourierAI/IDEA.
Accurate human posture classification in images and videos is crucial for automated applications across various fields, including work safety, physical rehabilitation, sports training, or daily assisted living. Recently, multimodal learning methods, such as Contrastive Language-Image Pretraining (CLIP), have advanced significantly in jointly understanding images and text. This study aims to assess the effectiveness of CLIP in classifying human postures, focusing on its application in yoga. Despite the initial limitations of the zero-shot approach, applying transfer learning on 15,301 images (real and synthetic) with 82 classes has shown promising results. The article describes the full procedure for fine-tuning, including the choice for image description syntax, models and hyperparameters adjustment. The fine-tuned CLIP model, tested on 3826 images, achieves an accuracy of over 85%, surpassing the current state-of-the-art of previous works on the same dataset by approximately 6%, its training time being 3.5 times lower than what is needed to fine-tune a YOLOv8-based model. For more application-oriented scenarios, with smaller datasets of six postures each, containing 1301 and 401 training images, the fine-tuned models attain an accuracy of 98.8% and 99.1%, respectively. Furthermore, our experiments indicate that training with as few as 20 images per pose can yield around 90% accuracy in a six-class dataset. This study demonstrates that this multimodal technique can be effectively used for yoga pose classification, and possibly for human posture classification, in general. Additionally, CLIP inference time (around 7 ms) supports that the model can be integrated into automated systems for posture evaluation, e.g., for developing a real-time personal yoga assistant for performance assessment.
The task of image captioning demands an algorithm to generate natural language descriptions of visual inputs. Recent advancements have seen a convergence between image captioning research and the development of Large Language Models (LLMs) and Multimodal LLMs -- like GPT-4V and Gemini -- which extend the capabilities of text-only LLMs to multiple modalities. This paper investigates whether Multimodal LLMs can supplant traditional image captioning networks by evaluating their performance on various image description benchmarks. We explore both the zero-shot capabilities of these models and their adaptability to different semantic domains through fine-tuning methods, including prompt learning, prefix tuning, and low-rank adaptation. Our results demonstrate that while Multimodal LLMs achieve impressive zero-shot performance, fine-tuning for specific domains while maintaining their generalization capabilities intact remains challenging. We discuss the implications of these findings for future research in image captioning and the development of more adaptable Multimodal LLMs.
Evaluating text-to-image generative models remains a challenge, despite the remarkable progress being made in their overall performances. While existing metrics like CLIPScore work for coarse evaluations, they lack the sensitivity to distinguish finer differences as model performance rapidly improves. In this work, we focus on the text rendering aspect of these models, which provides a lens for evaluating a generative model's fine-grained instruction-following capabilities. To this end, we introduce a new evaluation framework called TypeScore to sensitively assess a model's ability to generate images with high-fidelity embedded text by following precise instructions. We argue that this text generation capability serves as a proxy for general instruction-following ability in image synthesis. TypeScore uses an additional image description model and leverages an ensemble dissimilarity measure between the original and extracted text to evaluate the fidelity of the rendered text. Our proposed metric demonstrates greater resolution than CLIPScore to differentiate popular image generation models across a range of instructions with diverse text styles. Our study also evaluates how well these vision-language models (VLMs) adhere to stylistic instructions, disentangling style evaluation from embedded-text fidelity. Through human evaluation studies, we quantitatively meta-evaluate the effectiveness of the metric. Comprehensive analysis is conducted to explore factors such as text length, captioning models, and current progress towards human parity on this task. The framework provides insights into remaining gaps in instruction-following for image generation with embedded text.
Multimodal Chain of Thought (MCoT) is a popular prompting strategy for improving the performance of multimodal large language models (MLLMs) across a range of complex reasoning tasks. Despite its popularity, there is a notable absence of automated methods for evaluating the quality of reasoning steps in MCoT. To address this gap, we propose Multimodal Chain-of-Thought Evaluation (MiCEval), a framework designed to assess the correctness of reasoning chains by evaluating the quality of both the description and each reasoning step. The evaluation of the description component focuses on the accuracy of the image descriptions, while the reasoning step evaluates the quality of each step as it is conditionally generated based on the preceding steps. MiCEval is built upon a fine-grained dataset with annotations that rate each step according to correctness, relevance, and informativeness. Extensive experiments on four state-of-the-art MLLMs show that step-wise evaluations using MiCEval align more closely with human judgments compared to existing methods based on cosine similarity or fine-tuning approaches. MiCEval datasets and code can be found in https://github.com/alenai97/MiCEval.
Multimodal Large Language Models (MLLMs) demonstrate a strong understanding of the real world and can even handle complex tasks. However, they still fail on some straightforward visual question-answering (VQA) problems. This paper dives deeper into this issue, revealing that models tend to err when answering easy questions (e.g. Yes/No questions) about an image, even though they can correctly describe it. We refer to this model behavior discrepancy between difficult and simple questions as model laziness. To systematically investigate model laziness, we manually construct LazyBench, a benchmark that includes Yes/No, multiple choice, short answer questions, and image description tasks that are related to the same subjects in the images. Based on LazyBench, we observe that laziness widely exists in current advanced MLLMs (e.g. GPT-4o, Gemini-1.5-pro, Claude 3 and LLaVA-v1.5-13B), and it is more pronounced on stronger models. We also analyze the VQA v2 (LLaVA-v1.5-13B) benchmark and find that about half of its failure cases are caused by model laziness, which further highlights the importance of ensuring that the model fully utilizes its capability. To this end, we conduct preliminary exploration on how to mitigate laziness and find that chain of thought (CoT) can effectively address this issue.
Image description datasets play a crucial role in the advancement of various applications such as image understanding, text-to-image generation, and text-image retrieval. Currently, image description datasets primarily originate from two sources. One source is the scraping of image-text pairs from the web. Despite their abundance, these descriptions are often of low quality and noisy. Another is through human labeling. Datasets such as COCO are generally very short and lack details. Although detailed image descriptions can be annotated by humans, the high annotation cost limits the feasibility. These limitations underscore the need for more efficient and scalable methods to generate accurate and detailed image descriptions. In this paper, we propose an innovative framework termed Image Textualization (IT), which automatically produces high-quality image descriptions by leveraging existing multi-modal large language models (MLLMs) and multiple vision expert models in a collaborative manner, which maximally convert the visual information into text. To address the current lack of benchmarks for detailed descriptions, we propose several benchmarks for comprehensive evaluation, which verifies the quality of image descriptions created by our framework. Furthermore, we show that LLaVA-7B, benefiting from training on IT-curated descriptions, acquire improved capability to generate richer image descriptions, substantially increasing the length and detail of their output with less hallucination.
Vision-language models have emerged as a powerful tool for previously challenging multi-modal classification problem in the medical domain. This development has led to the exploration of automated image description generation for multi-modal clinical scans, particularly for radiology report generation. Existing research has focused on clinical descriptions for specific modalities or body regions, leaving a gap for a model providing entire-body multi-modal descriptions. In this paper, we address this gap by automating the generation of standardized body station(s) and list of organ(s) across the whole body in multi-modal MR and CT radiological images. Leveraging the versatility of the Contrastive Language-Image Pre-training (CLIP), we refine and augment the existing approach through multiple experiments, including baseline model fine-tuning, adding station(s) as a superset for better correlation between organs, along with image and language augmentations. Our proposed approach demonstrates 47.6% performance improvement over baseline PubMedCLIP.
Despite their significant advancements, Multimodal Large Language Models (MLLMs) often generate factually inaccurate information, referred to as hallucination. In this work, we address object hallucinations in MLLMs, where information is generated about an object not present in the input image. We introduce Data-augmented Phrase-level Alignment (DPA), a novel loss which can be applied to instruction-tuned off-the-shelf MLLMs to mitigate hallucinations, while preserving their general vision-language capabilities. To fine-tune MLLMs with DPA, we first generate a set of `hallucinated' and `correct' response pairs through generative data augmentation by selectively altering the ground-truth information of the correct responses at a phrase level. The DPA loss is then used to train MLLMs to reduce the likelihood of hallucinated phrases compared to the correct ones. Our thorough evaluation on various benchmarks confirms the effectiveness of DPA in mitigating hallucination while retaining the out-of-the-box performance of the MLLMs on general tasks. For instance, MLLMs finetuned with DPA, which we refer to as Hallucination Attenuated Language and Vision Assistant (HALVA), improve F1 by up to 13.4% on hallucination visual question-answering and reduce the hallucination rate by up to 4.2% on image description tasks.
Language has been useful in extending the vision encoder to data from diverse distributions without empirical discovery in training domains. However, as the image description is mostly at coarse-grained level and ignores visual details, the resulted embeddings are still ineffective in overcoming complexity of domains at inference time. We present a self-supervision framework WIDIn, Wording Images for Domain-Invariant representation, to disentangle discriminative visual representation, by only leveraging data in a single domain and without any test prior. Specifically, for each image, we first estimate the language embedding with fine-grained alignment, which can be consequently used to adaptively identify and then remove domain-specific counterpart from the raw visual embedding. WIDIn can be applied to both pretrained vision-language models like CLIP, and separately trained uni-modal models like MoCo and BERT. Experimental studies on three domain generalization datasets demonstrate the effectiveness of our approach.