Large vision-language models (LVLMs) have demonstrated impressive capabilities across diverse multimodal tasks, yet they remain highly susceptible to visual hallucinations (VH), often producing confident but inaccurate descriptions of visual content. Building on the insight that not all tokens and attention heads contribute equally to VH mitigation, we introduce VisFlow, a lightweight and training-free framework that alleviates hallucinations by directly modulating attention patterns during inference. To address two primary challenges of VH, namely insufficient visual attention and the dominance of language priors, we identify three problematic attention behaviors in LVLMs: (1) disproportionate allocation of attention to uninformative or trailing visual tokens, (2) over-dependence on the previously generated token, and (3) excessive fixation on system prompts that hinders multimodal integration. To overcome these issues, VisFlow introduces a dual-level Attention Intervention, consisting of Token-level Attention Intervention (TAI), which reinforces attention to salient visual regions, and Head-level Attention Intervention (HAI), which suppresses undue focus on system prompts and adjacent text tokens. Together, these interventions strengthen visual alignment while reducing linguistic bias. Extensive experiments across diverse models and benchmarks demonstrate that VisFlow effectively mitigates hallucinations with minimal computational overhead.

Given radiology images, automatic radiology report generation aims to produce informative text that reports diseases. It can benefit current clinical practice in diagnostic radiology. Existing methods typically rely on large-scale medical datasets annotated by clinicians to train desirable models. However, for novel diseases, sufficient training data are typically not available. We propose a prompt-based deep learning framework, i.e., PromptLLM, to align, autoencode, and prompt the (large) language model to generate reports for novel diseases accurately and efficiently. Our method includes three major steps: 1) aligning visual images and textual reports to learn general knowledge across modalities from diseases where labeled data are sufficient, 2) autoencoding the LLM using unlabeled data of novel diseases to learn the specific knowledge and writing styles of the novel disease, and 3) prompting the LLM with learned knowledge and writing styles to report the novel diseases contained in the radiology images. Through the above three steps, with limited labels on novel diseases, we show that PromptLLM can rapidly learn the corresponding knowledge for accurate novel disease reporting. The experiments on COVID-19 and diverse thorax diseases show that our approach, utilizing 1% of the training data, achieves desirable performance compared to previous methods. It shows that our approach allows us to relax the reliance on labeled data that is common to existing methods. It could have a real-world impact on data analysis during the early stages of novel diseases.

While vision-language pre-trained models (VL-PTMs) have advanced multimodal research in recent years, their mastery in a few languages like English restricts their applicability in broader communities. To this end, there is an increasing interest in developing multilingual VL models via a joint-learning setup, which, however, could be unrealistic due to expensive costs and data availability. In this work, we propose to extend VL-PTMs' language capacity by continual language learning (CLL), where a model needs to update its linguistic knowledge incrementally without suffering from catastrophic forgetting (CF). We begin our study by introducing a model dubbed CLL-CLIP, which builds upon CLIP, a prevailing VL-PTM that has acquired image-English text alignment. Specifically, CLL-CLIP contains an expandable token embedding layer to handle linguistic differences. It solely trains token embeddings to improve memory stability and is optimized under cross-modal and cross-lingual objectives to learn the alignment between images and multilingual texts. To alleviate CF raised by covariate shift and lexical overlap, we further propose a novel approach that ensures the identical distribution of all token embeddings during initialization and regularizes token embedding learning during training. We construct a CLL benchmark covering 36 languages based on MSCOCO and XM3600 datasets and then evaluate multilingual image-text retrieval performance. Extensive experiments verify the effectiveness of CLL-CLIP and show that our approach can boost CLL-CLIP, e.g., by 6.7% in text-to-image average Recall@1 on XM3600, and improve various state-of-the-art methods consistently. Our code and data are available at https://github.com/yangbang18/CLFM.

Medical report generation (MRG) is a challenging yet highly demanding task in the application of multi-modal artificial intelligence in medicine. Typically, training an MRG model requires tens of thousands of labelled radiology images and reports datasets, which could be impractical for most clinical research groups. In this study, we present C2RG, a novel 3D vision and language foundation model tailored for Coronary Computed Tomography Angiography (CTA) Report Generation. Inspired by BLIP-2’s architecture, our method integrates a self-supervised pre-trained 3D cardiac vision model (ViT-B) and a general-purpose bilingual foundation model (ChatGLM-6B), with a lightweight querying Transformer (Q-Former). We also introduce a parallel high-resolution feature extractor module and a coronary calcification evaluation loss to simultaneously encode fine-grained 3D features and constrain the accuracy of report generation. We compared our model with six state-of-the-art MRG methods on a clinical dataset with 118 subjects, comprising 453 paired 3D CTA images and radiology reports. Experimental results with extensive ablations show the efficacy of our C2RG. Codes will be open-sourced after the conference.

Continuous learning requires adapting quickly to incoming tasks while avoiding catastrophic forgetting. Typical solutions resort to a rehearsal buffer to replay old data, which is intractable to apply in real-world scenarios with limited memory and inaccessible privacy. Recently, with the emergence of large-scale pre-trained models, prompting methods have rapidly become a popular rehearsal-free alternative to rehearsal-based methods. The core of prmopting is to encode knowledge leveraging a set of parameters, however, knowledge decoupling and complementarity still remain some challenges. To tackle these challenges, this paper presents a KnowledgeComplementary Prompting approach, KC-Prompt, which end-to-end integrates and releases the task-invariant and task-specific knowledge for the ViT backbone. KC-Prompt designs knowledge maintenance and knowledge sharing mechanisms to form complementary prompt generators. In addition, we employ a components weighting method to instantiate prompt generators, making the training process fully differentiable. Sufficient experiments on CIFAR-100 and Split ImageNet-R benchmarks demonstrate the superiority of KC-Prompt in the challenging and realistic class-incremental learning setting.

Natural Language Generation (NLG) accepts input data in the form of images, videos, or text and generates corresponding natural language text as output. Existing NLG methods mainly adopt a supervised approach and rely heavily on coupled data-to-text pairs. However, for many targeted scenarios and for non-English languages, sufficient quantities of labeled data are often not available. As a result, it is necessary to collect and label data-text pairs for training, which is both costly and time-consuming. To relax the dependency on labeled data of downstream tasks, we propose an intuitive and effective zero-shot learning framework, ZeroNLG, which can deal with multiple NLG tasks, including image-to-text (image captioning), video-to-text (video captioning), and text-to-text (neural machine translation), across English, Chinese, German, and French within a unified framework. ZeroNLG does not require any labeled downstream pairs for training. During training, ZeroNLG (i) projects different domains (across modalities and languages) to corresponding coordinates in a shared common latent space; (ii) bridges different domains by aligning their corresponding coordinates in this space; and (iii) builds an unsupervised multilingual auto-encoder to learn to generate text by reconstructing the input text given its coordinate in shared latent space. Consequently, during inference, based on the data-to-text pipeline, ZeroNLG can generate target sentences across different languages given the coordinate of input data in the common space. Within this unified framework, given visual (imaging or video) data as input, ZeroNLG can perform zero-shot visual captioning; given textual sentences as input, ZeroNLG can perform zero-shot machine translation. We present the results of extensive experiments on twelve NLG tasks, showing that, without using any labeled downstream pairs for training, ZeroNLG generates high-quality and "believable" outputs and significantly outperforms existing zero-shot methods.

Zero-shot temporal action detection (ZS-TAD), aiming to recognize and detect new and unseen video actions, is an emerging and challenging task with limited solutions. Recent studies have adapted the vision-language pre-trained model CLIP for this task in a parameter-efficient fine-tuning fashion to achieve open-vocabulary detection. However, they suffer from insufficient vision-text alignment because of the dual-stream structure of CLIP and yield inferior TAD results due to the lack of accurate action prior. In this paper, we target the above limitations and propose to learn multimodal Prompts and Text-Enhanced Actionness (mProTEA) for ZS-TAD. Specifically, we insert learnable layer-wise prompts into the vision and text branches of the frozen CLIP and establish a strong coupling between them, resulting in multimodal prompts that can boost cross-modal alignment. To ease computation costs, we propose to conduct multimodal prompt learning on an image recognition dataset with rich concepts (e.g., ImageNet) first and then keep them frozen during TAD fine-tuning. For improving TAD, we introduce text-enhanced actionness modeling, where we leverage the concise semantics of text to assist the calculation of class-agnostic actionness scores, to offer accurate prior information for both action classification and localization. With the above designs, our mProTEA excels in extensive TAD experiments, surpassing the strong competitor STALE by 5.1% on ActivityNet under the zero-shot setting and achieving state-of-the-art performance in conventional supervised scenarios. Ablation studies confirm the effectiveness of our proposals and show superior domain generalization of multimodal prompts learned on ImageNet against the other 10 image recognition datasets.

Supervised visual captioning models typically require a large scale of images or videos paired with descriptions in a specific language (i.e., the vision-caption pairs) for training.However, collecting and labeling large-scale datasets is time-consuming and expensive for many scenarios and languages.Therefore, sufficient labeled pairs are usually not available.To deal with the label shortage problem, we present a simple yet effective zero-shot approach Mul-tiCapCLIP that can generate visual captions for different scenarios and languages without any labeled vision-caption pairs of downstream datasets.In the training stage, MultiCapCLIP only requires text data for input.Then it conducts two main steps: 1) retrieving concept prompts that preserve the corresponding domain knowledge of new scenarios; 2) autoencoding the prompts to learn writing styles to output captions in a desired language.In the testing stage, MultiCapCLIP instead takes visual data as input directly to retrieve the concept prompts to generate the final visual descriptions.The extensive experiments on image and video captioning across four benchmarks and four languages (i.e., English, Chinese, German, and French) confirm the effectiveness of our approach.Compared with state-of-theart zero-shot and weakly-supervised methods, our method achieves 4.8% and 21.5% absolute improvements in terms of BLEU@4 and CIDEr metrics.Our code is available at https: //github.com/yangbang18/MultiCapCLIP.

Automatic radiology report generation has attracted enormous research interest due to its practical value in reducing the workload of radiologists. However, simultaneously establishing global correspondences between the image (e.g., Chest X-ray) and its related report and local alignments between image patches and keywords remains challenging. To this end, we propose an Unify, Align and then Refine (UAR) approach to learn multi-level cross-modal alignments and introduce three novel modules: Latent Space Unifier (LSU), Cross-modal Representation Aligner (CRA) and Text-to-Image Refiner (TIR). Specifically, LSU unifies multimodal data into discrete tokens, making it flexible to learn common knowledge among modalities with a shared network. The modality-agnostic CRA learns discriminative features via a set of orthonormal basis and a dual-gate mechanism first and then globally aligns visual and textual representations under a triplet contrastive loss. TIR boosts token-level local alignment via calibrating text-to-image attention with a learnable mask. Additionally, we design a two-stage training procedure to make UAR gradually grasp cross-modal alignments at different levels, which imitates radiologists’ workflow: writing sentence by sentence first and then checking word by word. Extensive experiments and analyses on IU-Xray and MIMIC-CXR benchmark datasets demonstrate the superiority of our UAR against varied state-of-the-art methods.

Concepts, a collective term for meaningful words that correspond to objects, actions, and attributes, can act as an intermediary for video captioning. While many efforts have been made to augment video captioning with concepts, most methods suffer from limited precision of concept detection and insufficient utilization of concepts, which could provide caption generation with inaccurate and inadequate prior information. Considering these issues, we propose a Concept-awARE video captioning framework (CARE) to facilitate plausible caption generation. Based on the encoder-decoder structure, CARE detects concepts precisely via multimodal-driven concept detection (MCD) and offers sufficient prior information to caption generation by global-local semantic guidance (G-LSG). Specifically, we implement MCD by leveraging video-to-text retrieval and the multimedia nature of videos. To achieve G-LSG, given the concept probabilities predicted by MCD, we weight and aggregate concepts to mine the video's latent topic to affect decoding globally and devise a simple yet efficient hybrid attention module to exploit concepts and video content to impact decoding locally. Finally, to develop CARE, we emphasize on the knowledge transfer of a contrastive vision-language pre-trained model (i.e., CLIP) in terms of visual understanding and video-to-text retrieval. With the multi-role CLIP, CARE can outperform CLIP-based strong video captioning baselines with affordable extra parameter and inference latency costs. Extensive experiments on MSVD, MSR-VTT, and VATEX datasets demonstrate the versatility of our approach for different encoder-decoder networks and the superiority of CARE against state-of-the-art methods. Our code is available at https://github.com/yangbang18/CARE.

Gene Ontology (GO) is the primary gene function knowledge base that enables computational tasks in biomedicine. The basic element of GO is a term, which includes a set of genes with the same function. Existing research efforts of GO mainly focus on predicting gene term associations. Other tasks, such as generating descriptions of new terms, are rarely pursued. In this paper, we propose a novel task: GO term description generation. This task aims to automatically generate a sentence that describes the function of a GO term belonging to one of the three categories, i.e., molecular function, biological process, and cellular component. To address this task, we propose a Graph-in-Graph network that can efficiently leverage the structural information of GO. The proposed network introduces a two-layer graph: the first layer is a graph of GO terms where each node is also a graph (gene graph). Such a Graph-in-Graph network can derive the biological functions of GO terms and generate proper descriptions. To validate the effectiveness of the proposed network, we build three large-scale benchmark datasets. By incorporating the proposed Graph-in-Graph network, the performances of seven different sequence-to-sequence models can be substantially boosted across all evaluation metrics, with up to 34.7%, 14.5%, and 39.1% relative improvements in BLEU, ROUGE-L, and METEOR, respectively.

The "Patient Instruction" (PI), which contains critical instructional information provided both to carers and to the patient at the time of discharge, is essential for the patient to manage their condition outside hospital. An accurate and easy-to-follow PI can improve the self-management of patients which can in turn reduce hospital readmission rates. However, writing an appropriate PI can be extremely time consuming for physicians, and is subject to being incomplete or error-prone for (potentially overworked) physicians. Therefore, we propose a new task that can provide an objective means of avoiding incompleteness, while reducing clinical workload: the automatic generation of the PI, which is imagined as being a document that the clinician can review, modify, and approve as necessary (rather than taking the human "out of the loop"). We build a benchmark clinical dataset and propose the Re$^3$Writer, which imitates the working patterns of physicians to first retrieve related working experience from historical PIs written by physicians, then reason related medical knowledge. Finally, it refines the retrieved working experience and reasoned medical knowledge to extract useful information, which is used to generate the PI for previously-unseen patient according to their health records during hospitalization. Our experiments show that, using our method, the performance of 6 different models can be substantially boosted across all metrics, with up to 20%, 11%, and 19% relative improvements in BLEU-4, ROUGE-L, and METEOR, respectively. Meanwhile, we show results from human evaluations to measure the effectiveness in terms of its usefulness for clinical practice. The code is available at https://github.com/AI-in-Health/Patient-Instructions.

It is encouraged to see that progress has been made to bridge videos and natural language. However, mainstream video captioning methods suffer from slow inference speed due to the sequential manner of autoregressive decoding, and prefer generating generic descriptions due to the insufficient training of visual words (e.g., nouns and verbs) and inadequate decoding paradigm. In this paper, we propose a non-autoregressive decoding based model with a coarse-to-fine captioning procedure to alleviate these defects. In implementations, we employ a bi-directional self-attention based network as our language model for achieving inference speedup, based on which we decompose the captioning procedure into two stages, where the model has different focuses. Specifically, given that visual words determine the semantic correctness of captions, we design a mechanism of generating visual words to not only promote the training of scene-related words but also capture relevant details from videos to construct a coarse-grained sentence ``template''. Thereafter, we devise dedicated decoding algorithms that fill in the ``template'' with suitable words and modify inappropriate phrasing via iterative refinement to obtain a fine-grained description. Extensive experiments on two mainstream video captioning benchmarks, i.e., MSVD and MSR-VTT, demonstrate that our approach achieves state-of-the-art performance, generates diverse descriptions, and obtains high inference efficiency.

Video captioning is a challenging task which aims at automatically generating a natural language description of a given video. Recent researches have shown that exploiting the intrinsic multi-modalities of videos significantly promotes captioning performance. However, how to integrate multi-modalities to generate effective semantic representations for video captioning is still an open issue. Some researchers proposed to learn multimodal features in parallel during the encoding stage. The downside of these methods lies in the neglect of the interaction among multi-modalities and their rich contextual information. In this study, inspired by the fact that visual contents are generally more important for comprehending videos, we propose a novel Visual Oriented Encoder (VOE) to integrate multimodal features in an interactive manner. Specifically, VOE is designed as a hierarchical structure, where bottom layers are utilized to extract multi-scale contexts from auxiliary modalities while the top layer is exploited to generate joint representations by considering both visual and contextual information. Following the encoder-decoder framework, we systematically develop a VOE-LSTM model and evaluate it on two mainstream benchmarks: MSVD and MSR-VTT. Experimental results show that the proposed VOE surpasses conventional encoders and our VOE-LSTM model achieves competitive results compared with state-of-the-art approaches.