Electronic Health Records (EHRs) hold detailed longitudinal information about each patient's health status and general clinical history, a large portion of which is stored within the unstructured text. Temporal modelling of this medical history, which considers the sequence of events, can be used to forecast and simulate future events, estimate risk, suggest alternative diagnoses or forecast complications. While most prediction approaches use mainly structured data or a subset of single-domain forecasts and outcomes, we processed the entire free-text portion of EHRs for longitudinal modelling. We present Foresight, a novel GPT3-based pipeline that uses NER+L tools (i.e. MedCAT) to convert document text into structured, coded concepts, followed by providing probabilistic forecasts for future medical events such as disorders, medications, symptoms and interventions. Since large portions of EHR data are in text form, such an approach benefits from a granular and detailed view of a patient while introducing modest additional noise. On tests in two large UK hospitals (King's College Hospital, South London and Maudsley) and the US MIMIC-III dataset precision@10 of 0.80, 0.81 and 0.91 was achieved for forecasting the next biomedical concept. Foresight was also validated on 34 synthetic patient timelines by 5 clinicians and achieved relevancy of 97% for the top forecasted candidate disorder. Foresight can be easily trained and deployed locally as it only requires free-text data (as a minimum). As a generative model, it can simulate follow-on disorders, medications and interventions for as many steps as required. Foresight is a general-purpose model for biomedical concept modelling that can be used for real-world risk estimation, virtual trials and clinical research to study the progression of diseases, simulate interventions and counterfactuals, and for educational purposes.
Cross-modal fashion image synthesis has emerged as one of the most promising directions in the generation domain due to the vast untapped potential of incorporating multiple modalities and the wide range of fashion image applications. To facilitate accurate generation, cross-modal synthesis methods typically rely on Contrastive Language-Image Pre-training (CLIP) to align textual and garment information. In this work, we argue that simply aligning texture and garment information is not sufficient to capture the semantics of the visual information and therefore propose MaskCLIP. MaskCLIP decomposes the garments into semantic parts, ensuring fine-grained and semantically accurate alignment between the visual and text information. Building on MaskCLIP, we propose ARMANI, a unified cross-modal fashion designer with part-level garment-text alignment. ARMANI discretizes an image into uniform tokens based on a learned cross-modal codebook in its first stage and uses a Transformer to model the distribution of image tokens for a real image given the tokens of the control signals in its second stage. Contrary to prior approaches that also rely on two-stage paradigms, ARMANI introduces textual tokens into the codebook, making it possible for the model to utilize fine-grain semantic information to generate more realistic images. Further, by introducing a cross-modal Transformer, ARMANI is versatile and can accomplish image synthesis from various control signals, such as pure text, sketch images, and partial images. Extensive experiments conducted on our newly collected cross-modal fashion dataset demonstrate that ARMANI generates photo-realistic images in diverse synthesis tasks and outperforms existing state-of-the-art cross-modal image synthesis approaches.Our code is available at https://github.com/Harvey594/ARMANI.
Contrastive representation learning has gained much attention due to its superior performance in learning representations from both image and sequential data. However, the learned representations could potentially lead to performance disparities in downstream tasks, such as increased silencing of underrepresented groups in toxicity comment classification. In light of this challenge, in this work, we study learning fair representations that satisfy a notion of fairness known as equalized odds for text classification via contrastive learning. Specifically, we first theoretically analyze the connections between learning representations with fairness constraint and conditional supervised contrastive objectives. Inspired by our theoretical findings, we propose to use conditional supervised contrastive objectives to learn fair representations for text classification. We conduct experiments on two text datasets to demonstrate the effectiveness of our approaches in balancing the trade-offs between task performance and bias mitigation among existing baselines for text classification. Furthermore, we also show that the proposed methods are stable in different hyperparameter settings.
The cost of vision-and-language pre-training has become increasingly prohibitive due to end-to-end training of large-scale models. This paper proposes BLIP-2, a generic and efficient pre-training strategy that bootstraps vision-language pre-training from off-the-shelf frozen pre-trained image encoders and frozen large language models. BLIP-2 bridges the modality gap with a lightweight Querying Transformer, which is pre-trained in two stages. The first stage bootstraps vision-language representation learning from a frozen image encoder. The second stage bootstraps vision-to-language generative learning from a frozen language model. BLIP-2 achieves state-of-the-art performance on various vision-language tasks, despite having significantly fewer trainable parameters than existing methods. For example, our model outperforms Flamingo80B by 8.7% on zero-shot VQAv2 with 54x fewer trainable parameters. We also demonstrate the model's emerging capabilities of zero-shot image-to-text generation that can follow natural language instructions.
One-shot voice conversion (VC) aims to convert speech from any source speaker to an arbitrary target speaker with only a few seconds of reference speech from the target speaker. This relies heavily on disentangling the speaker's identity and speech content, a task that still remains challenging. Here, we propose a novel approach to learning disentangled speech representation by transfer learning from style-based text-to-speech (TTS) models. With cycle consistent and adversarial training, the style-based TTS models can perform transcription-guided one-shot VC with high fidelity and similarity. By learning an additional mel-spectrogram encoder through a teacher-student knowledge transfer and novel data augmentation scheme, our approach results in disentangled speech representation without needing the input text. The subjective evaluation shows that our approach can significantly outperform the previous state-of-the-art one-shot voice conversion models in both naturalness and similarity.
Large pre-trained language models are capable of generating varied and fluent texts. Starting from the prompt, these models generate a narrative that can develop unpredictably. The existing methods of controllable text generation, which guide the narrative in the text in the user-specified direction, require creating a training corpus and an additional time-consuming training procedure. The paper proposes and investigates Collocation2Text, a plug-and-play method for automatic controllable text generation in Russian, which does not require fine-tuning. The method is based on two interacting models: the autoregressive language ruGPT-3 model and the autoencoding language ruRoBERTa model. The idea of the method is to shift the output distribution of the autoregressive model according to the output distribution of the autoencoding model in order to ensure a coherent transition of the narrative in the text towards the guide phrase, which can contain single words or collocations. The autoencoding model, which is able to take into account the left and right contexts of the token, "tells" the autoregressive model which tokens are the most and least logical at the current generation step, increasing or decreasing the probabilities of the corresponding tokens. The experiments on generating news articles using the proposed method showed its effectiveness for automatically generated fluent texts which contain coherent transitions between user-specified phrases.
Diffusion-based generative models have shown great potential for image synthesis, but there is a lack of research on the security and privacy risks they may pose. In this paper, we investigate the vulnerability of diffusion models to Membership Inference Attacks (MIAs), a common privacy concern. Our results indicate that existing MIAs designed for GANs or VAE are largely ineffective on diffusion models, either due to inapplicable scenarios (e.g., requiring the discriminator of GANs) or inappropriate assumptions (e.g., closer distances between synthetic images and member images). To address this gap, we propose Step-wise Error Comparing Membership Inference (SecMI), a black-box MIA that infers memberships by assessing the matching of forward process posterior estimation at each timestep. SecMI follows the common overfitting assumption in MIA where member samples normally have smaller estimation errors, compared with hold-out samples. We consider both the standard diffusion models, e.g., DDPM, and the text-to-image diffusion models, e.g., Stable Diffusion. Experimental results demonstrate that our methods precisely infer the membership with high confidence on both of the two scenarios across six different datasets
In this work we explore how language models can be employed to analyze language and discriminate between mentally impaired and healthy subjects through the perplexity metric. Perplexity was originally conceived as an information-theoretic measure to assess how much a given language model is suited to predict a text sequence or, equivalently, how much a word sequence fits into a specific language model. We carried out an extensive experimentation with the publicly available data, and employed language models as diverse as N-grams, from 2-grams to 5-grams, and GPT-2, a transformer-based language model. We investigated whether perplexity scores may be used to discriminate between the transcripts of healthy subjects and subjects suffering from Alzheimer Disease (AD). Our best performing models achieved full accuracy and F-score (1.00 in both precision/specificity and recall/sensitivity) in categorizing subjects from both the AD class and control subjects. These results suggest that perplexity can be a valuable analytical metrics with potential application to supporting early diagnosis of symptoms of mental disorders.
Self-supervised auto-encoders have emerged as a successful framework for representation learning in computer vision and natural language processing in recent years, However, their application to graph data has been met with limited performance due to the non-Euclidean and complex structure of graphs in comparison to images or text, as well as the limitations of conventional auto-encoder architectures. In this paper, we investigate factors impacting the performance of auto-encoders on graph data and propose a novel auto-encoder model for graph representation learning. Our model incorporates a hierarchical adaptive masking mechanism to incrementally increase the difficulty of training in order to mimic the process of human cognitive learning, and a trainable corruption scheme to enhance the robustness of learned representations. Through extensive experimentation on ten benchmark datasets, we demonstrate the superiority of our proposed method over state-of-the-art graph representation learning models.
The task of converting text input into video content is becoming an important topic for synthetic media generation. Several methods have been proposed with some of them reaching close-to-natural performances in constrained tasks. In this paper, we tackle a subissue of the text-to-video generation problem, by converting the text into lip landmarks. However, we do this using a modular, controllable system architecture and evaluate each of its individual components. Our system, entitled FlexLip, is split into two separate modules: text-to-speech and speech-to-lip, both having underlying controllable deep neural network architectures. This modularity enables the easy replacement of each of its components, while also ensuring the fast adaptation to new speaker identities by disentangling or projecting the input features. We show that by using as little as 20 min of data for the audio generation component, and as little as 5 min for the speech-to-lip component, the objective measures of the generated lip landmarks are comparable with those obtained when using a larger set of training samples. We also introduce a series of objective evaluation measures over the complete flow of our system by taking into consideration several aspects of the data and system configuration. These aspects pertain to the quality and amount of training data, the use of pretrained models, and the data contained therein, as well as the identity of the target speaker; with regard to the latter, we show that we can perform zero-shot lip adaptation to an unseen identity by simply updating the shape of the lips in our model.