We introduce the Song Describer dataset (SDD), a new crowdsourced corpus of high-quality audio-caption pairs, designed for the evaluation of music-and-language models. The dataset consists of 1.1k human-written natural language descriptions of 706 music recordings, all publicly accessible and released under Creative Common licenses. To showcase the use of our dataset, we benchmark popular models on three key music-and-language tasks (music captioning, text-to-music generation and music-language retrieval). Our experiments highlight the importance of cross-dataset evaluation and offer insights into how researchers can use SDD to gain a broader understanding of model performance.
CLIP is a widely used foundational vision-language model that is used for zero-shot image recognition and other image-text alignment tasks. We demonstrate that CLIP is vulnerable to change in image quality under compression. This surprising result is further analysed using an attribution method-Integrated Gradients. Using this attribution method, we are able to better understand both quantitatively and qualitatively exactly the nature in which the compression affects the zero-shot recognition accuracy of this model. We evaluate this extensively on CIFAR-10 and STL-10. Our work provides the basis to understand this vulnerability of CLIP and can help us develop more effective methods to improve the robustness of CLIP and other vision-language models.
Most pretrained language models rely on subword tokenization, which processes text as a sequence of subword tokens. However, different granularities of text, such as characters, subwords, and words, can contain different kinds of information. Previous studies have shown that incorporating multiple input granularities improves model generalization, yet very few of them outputs useful representations for each granularity. In this paper, we introduce the entanglement model, aiming to combine character and subword language models. Inspired by vision-language models, our model treats characters and subwords as separate modalities, and it generates mutually informed representations for both granularities as output. We evaluate our model on text classification, named entity recognition, and POS-tagging tasks. Notably, the entanglement model outperforms its backbone language models, particularly in the presence of noisy texts and low-resource languages. Furthermore, the entanglement model even outperforms larger pre-trained models on all English sequence labeling tasks and classification tasks. Our anonymized code is available at https://anonymous.4open.science/r/noisy-IE-A673
Empathy is a social skill that indicates an individual's ability to understand others. Over the past few years, empathy has drawn attention from various disciplines, including but not limited to Affective Computing, Cognitive Science and Psychology. Empathy is a context-dependent term; thus, detecting or recognising empathy has potential applications in society, healthcare and education. Despite being a broad and overlapping topic, the avenue of empathy detection studies leveraging Machine Learning remains underexplored from a holistic literature perspective. To this end, we systematically collect and screen 801 papers from 10 well-known databases and analyse the selected 54 papers. We group the papers based on input modalities of empathy detection systems, i.e., text, audiovisual, audio and physiological signals. We examine modality-specific pre-processing and network architecture design protocols, popular dataset descriptions and availability details, and evaluation protocols. We further discuss the potential applications, deployment challenges and research gaps in the Affective Computing-based empathy domain, which can facilitate new avenues of exploration. We believe that our work is a stepping stone to developing a privacy-preserving and unbiased empathic system inclusive of culture, diversity and multilingualism that can be deployed in practice to enhance the overall well-being of human life.
Diffusion-based generative models have significantly advanced text-to-image generation but encounter challenges when processing lengthy and intricate text prompts describing complex scenes with multiple objects. While excelling in generating images from short, single-object descriptions, these models often struggle to faithfully capture all the nuanced details within longer and more elaborate textual inputs. In response, we present a novel approach leveraging Large Language Models (LLMs) to extract critical components from text prompts, including bounding box coordinates for foreground objects, detailed textual descriptions for individual objects, and a succinct background context. These components form the foundation of our layout-to-image generation model, which operates in two phases. The initial Global Scene Generation utilizes object layouts and background context to create an initial scene but often falls short in faithfully representing object characteristics as specified in the prompts. To address this limitation, we introduce an Iterative Refinement Scheme that iteratively evaluates and refines box-level content to align them with their textual descriptions, recomposing objects as needed to ensure consistency. Our evaluation on complex prompts featuring multiple objects demonstrates a substantial improvement in recall compared to baseline diffusion models. This is further validated by a user study, underscoring the efficacy of our approach in generating coherent and detailed scenes from intricate textual inputs.
Medical image interpretation is central to most clinical applications such as disease diagnosis, treatment planning, and prognostication. In clinical practice, radiologists examine medical images and manually compile their findings into reports, which can be a time-consuming process. Automated approaches to radiology report generation, therefore, can reduce radiologist workload and improve efficiency in the clinical pathway. While recent deep-learning approaches for automated report generation from medical images have seen some success, most studies have relied on image-derived features alone, ignoring non-imaging patient data. Although a few studies have included the word-level contexts along with the image, the use of patient demographics is still unexplored. This paper proposes a novel multi-modal transformer network that integrates chest x-ray (CXR) images and associated patient demographic information, to synthesise patient-specific radiology reports. The proposed network uses a convolutional neural network to extract visual features from CXRs and a transformer-based encoder-decoder network that combines the visual features with semantic text embeddings of patient demographic information, to synthesise full-text radiology reports. Data from two public databases were used to train and evaluate the proposed approach. CXRs and reports were extracted from the MIMIC-CXR database and combined with corresponding patients' data MIMIC-IV. Based on the evaluation metrics used including patient demographic information was found to improve the quality of reports generated using the proposed approach, relative to a baseline network trained using CXRs alone. The proposed approach shows potential for enhancing radiology report generation by leveraging rich patient metadata and combining semantic text embeddings derived thereof, with medical image-derived visual features.
Bias analysis is a crucial step in the process of creating fair datasets for training and evaluating computer vision models. The bottleneck in dataset analysis is annotation, which typically requires: (1) specifying a list of attributes relevant to the dataset domain, and (2) classifying each image-attribute pair. While the second step has made rapid progress in automation, the first has remained human-centered, requiring an experimenter to compile lists of in-domain attributes. However, an experimenter may have limited foresight leading to annotation "blind spots," which in turn can lead to flawed downstream dataset analyses. To combat this, we propose GELDA, a nearly automatic framework that leverages large generative language models (LLMs) to propose and label various attributes for a domain. GELDA takes a user-defined domain caption (e.g., "a photo of a bird," "a photo of a living room") and uses an LLM to hierarchically generate attributes. In addition, GELDA uses the LLM to decide which of a set of vision-language models (VLMs) to use to classify each attribute in images. Results on real datasets show that GELDA can generate accurate and diverse visual attribute suggestions, and uncover biases such as confounding between class labels and background features. Results on synthetic datasets demonstrate that GELDA can be used to evaluate the biases of text-to-image diffusion models and generative adversarial networks. Overall, we show that while GELDA is not accurate enough to replace human annotators, it can serve as a complementary tool to help humans analyze datasets in a cheap, low-effort, and flexible manner.
Identity-consistent video generation seeks to synthesize videos that are guided by both textual prompts and reference images of entities. Current approaches typically utilize cross-attention layers to integrate the appearance of the entity, which predominantly captures semantic attributes, resulting in compromised fidelity of entities. Moreover, these methods necessitate iterative fine-tuning for each new entity encountered, thereby limiting their applicability. To address these challenges, we introduce VideoAssembler, a novel end-to-end framework for identity-consistent video generation that can conduct inference directly when encountering new entities. VideoAssembler is adept at producing videos that are not only flexible with respect to the input reference entities but also responsive to textual conditions. Additionally, by modulating the quantity of input images for the entity, VideoAssembler enables the execution of tasks ranging from image-to-video generation to sophisticated video editing. VideoAssembler comprises two principal components: the Reference Entity Pyramid (REP) encoder and the Entity-Prompt Attention Fusion (EPAF) module. The REP encoder is designed to infuse comprehensive appearance details into the denoising stages of the stable diffusion model. Concurrently, the EPAF module is utilized to integrate text-aligned features effectively. Furthermore, to mitigate the challenge of scarce data, we present a methodology for the preprocessing of training data. Our evaluation of the VideoAssembler framework on the UCF-101, MSR-VTT, and DAVIS datasets indicates that it achieves good performances in both quantitative and qualitative analyses (346.84 in FVD and 48.01 in IS on UCF-101). Our project page is at https://videoassembler.github.io/videoassembler.
Recent advancements in text-to-image models, particularly diffusion models, have shown significant promise. However, compositional text-to-image models frequently encounter difficulties in generating high-quality images that accurately align with input texts describing multiple objects, variable attributes, and intricate spatial relationships. To address this limitation, we employ large vision-language models (LVLMs) for multi-dimensional assessment of the alignment between generated images and their corresponding input texts. Utilizing this assessment, we fine-tune the diffusion model to enhance its alignment capabilities. During the inference phase, an initial image is produced using the fine-tuned diffusion model. The LVLM is then employed to pinpoint areas of misalignment in the initial image, which are subsequently corrected using the image editing algorithm until no further misalignments are detected by the LVLM. The resultant image is consequently more closely aligned with the input text. Our experimental results validate that the proposed methodology significantly improves text-image alignment in compositional image generation, particularly with respect to object number, attribute binding, spatial relationships, and aesthetic quality.
Text-to-video editing aims to edit the visual appearance of a source video conditional on textual prompts. A major challenge in this task is to ensure that all frames in the edited video are visually consistent. Most recent works apply advanced text-to-image diffusion models to this task by inflating 2D spatial attention in the U-Net into spatio-temporal attention. Although temporal context can be added through spatio-temporal attention, it may introduce some irrelevant information for each patch and therefore cause inconsistency in the edited video. In this paper, for the first time, we introduce optical flow into the attention module in the diffusion model's U-Net to address the inconsistency issue for text-to-video editing. Our method, FLATTEN, enforces the patches on the same flow path across different frames to attend to each other in the attention module, thus improving the visual consistency in the edited videos. Additionally, our method is training-free and can be seamlessly integrated into any diffusion-based text-to-video editing methods and improve their visual consistency. Experiment results on existing text-to-video editing benchmarks show that our proposed method achieves the new state-of-the-art performance. In particular, our method excels in maintaining the visual consistency in the edited videos.