Previous Sign Language Translation (SLT) methods achieve superior performance by relying on gloss annotations. However, labeling high-quality glosses is a labor-intensive task, which limits the further development of SLT. Although some approaches work towards gloss-free SLT through jointly training the visual encoder and translation network, these efforts still suffer from poor performance and inefficient use of the powerful Large Language Model (LLM). Most seriously, we find that directly introducing LLM into SLT will lead to insufficient learning of visual representations as LLM dominates the learning curve. To address these problems, we propose Factorized Learning assisted with Large Language Model (FLa-LLM) for gloss-free SLT. Concretely, we factorize the training process into two stages. In the visual initialing stage, we employ a lightweight translation model after the visual encoder to pre-train the visual encoder. In the LLM fine-tuning stage, we freeze the acquired knowledge in the visual encoder and integrate it with a pre-trained LLM to inspire the LLM’s translation potential. This factorized training strategy proves to be highly effective as evidenced by significant improvements achieved across three SLT datasets which are all conducted under the gloss-free setting.

This paper describes the NPU-MSXF system for the IWSLT 2023 speech-to-speech translation (S2ST) task which aims to translate from English speech of multi-source to Chinese speech. The system is built in a cascaded manner consisting of automatic speech recognition (ASR), machine translation (MT), and text-to-speech (TTS). We make tremendous efforts to handle the challenging multi-source input. Specifically, to improve the robustness to multi-source speech input, we adopt various data augmentation strategies and a ROVER-based score fusion on multiple ASR model outputs. To better handle the noisy ASR transcripts, we introduce a three-stage fine-tuning strategy to improve translation accuracy. Finally, we build a TTS model with high naturalness and sound quality, which leverages a two-stage framework, using network bottleneck features as a robust intermediate representation for speaker timbre and linguistic content disentanglement. Based on the two-stage framework, pre-trained speaker embedding is leveraged as a condition to transfer the speaker timbre in the source English speech to the translated Chinese speech. Experimental results show that our system has high translation accuracy, speech naturalness, sound quality, and speaker similarity. Moreover, it shows good robustness to multi-source data.

Clinical outcome prediction is critical to the condition prediction of patients and management of hospital capacities. There are two kinds of medical data, including time series signals recorded by various devices and clinical notes in electronic health records (EHR), which are used for two common prediction targets: mortality and length of stay. Traditional methods focused on utilizing time series data but ignored clinical notes. With the development of deep learning, natural language processing (NLP) and multi-modal learning methods are exploited to jointly model the time series and clinical notes with different modals. However, the existing methods failed to fuse the multi-modal features of patients from different views. Therefore, we propose the patient multi-view multi-modal feature fusion networks for clinical outcome prediction. Firstly, from patient inner view, we propose to utilize the co-attention module to enhance the fine-grained feature interaction between time series and clinical notes from each patient. Secondly, the patient outer view is the correlation between patients, which can be reflected by the structural knowledge in clinical notes. We exploit the structural information extracted from clinical notes to construct the patient correlation graph, and fuse patients’ multi-modal features by graph neural networks (GNN). The experimental results on MIMIC-III benchmark demonstrate the superiority of our method.

Multimodal named entity recognition (MNER) on social media is a challenging task which aims to extract named entities in free text and incorporate images to classify them into user-defined types. However, the annotation for named entities on social media demands a mount of human efforts. The existing semi-supervised named entity recognition methods focus on the text modal and are utilized to reduce labeling costs in traditional NER. However, the previous methods are not efficient for semi-supervised MNER. Because the MNER task is defined to combine the text information with image one and needs to consider the mismatch between the posted text and image. To fuse the text and image features for MNER effectively under semi-supervised setting, we propose a novel span-based multimodal variational autoencoder (SMVAE) model for semi-supervised MNER. The proposed method exploits modal-specific VAEs to model text and image latent features, and utilizes product-of-experts to acquire multimodal features. In our approach, the implicit relations between labels and multimodal features are modeled by multimodal VAE. Thus, the useful information of unlabeled data can be exploited in our method under semi-supervised setting. Experimental results on two benchmark datasets demonstrate that our approach not only outperforms baselines under supervised setting, but also improves MNER performance with less labeled data than existing semi-supervised methods.

Multimodal knowledge graph completion (MKGC) aims to predict missing entities in MKGs. Previous works usually share relation representation across modalities. This results in mutual interference between modalities during training, since for a pair of entities, the relation from one modality probably contradicts that from another modality. Furthermore, making a unified prediction based on the shared relation representation treats the input in different modalities equally, while their importance to the MKGC task should be different. In this paper, we propose MoSE, a Modality Split representation learning and Ensemble inference framework for MKGC. Specifically, in the training phase, we learn modality-split relation embeddings for each modality instead of a single modality-shared one, which alleviates the modality interference. Based on these embeddings, in the inference phase, we first make modality-split predictions and then exploit various ensemble methods to combine the predictions with different weights, which models the modality importance dynamically. Experimental results on three KG datasets show that MoSE outperforms state-of-the-art MKGC methods. Codes are available at https://github.com/OreOZhao/MoSE4MKGC.

Entity linking, which aims at aligning ambiguous entity mentions to their referent entities in a knowledge base, plays a key role in multiple natural language processing tasks. Recently, zero-shot entity linking task has become a research hotspot, which links mentions to unseen entities to challenge the generalization ability. For this task, the training set and test set are from different domains, and thus entity linking models tend to be overfitting due to the tendency of memorizing the properties of entities that appear frequently in the training set. We argue that general ultra-fine-grained type information can help the linking models to learn contextual commonality and improve their generalization ability to tackle the overfitting problem. However, in the zero-shot entity linking setting, any type information is not available and entities are only identified by textual descriptions. Thus, we first extract the ultra-fine entity type information from the entity textual descriptions. Then, we propose a hierarchical multi-task model to improve the high-level zero-shot entity linking candidate generation task by utilizing the entity typing task as an auxiliary low-level task, which introduces extracted ultra-fine type information into the candidate generation task. Experimental results demonstrate the effectiveness of utilizing the ultra-fine entity type information and our proposed method achieves state-of-the-art performance.

ICD coding is designed to assign the disease codes to electronic health records (EHRs) upon discharge, which is crucial for billing and clinical statistics. In an attempt to improve the effectiveness and efficiency of manual coding, many methods have been proposed to automatically predict ICD codes from clinical notes. However, most previous works ignore the decisive information contained in structured medical data in EHRs, which is hard to be captured from the noisy clinical notes. In this paper, we propose a Tree-enhanced Multimodal Attention Network (TreeMAN) to fuse tabular features and textual features into multimodal representations by enhancing the text representations with tree-based features via the attention mechanism. Tree-based features are constructed according to decision trees learned from structured multimodal medical data, which capture the decisive information about ICD coding. We can apply the same multi-label classifier from previous text models to the multimodal representations to predict ICD codes. Experiments on two MIMIC datasets show that our method outperforms prior state-of-the-art ICD coding approaches. The code is available at https://github.com/liu-zichen/TreeMAN.

Despite the great progress of Visual Question Answering (VQA), current VQA models heavily rely on the superficial correlation between the question type and its corresponding frequent answers (i.e., language priors) to make predictions, without really understanding the input. In this work, we define the training instances with the same question type but different answers as superficially similar instances, and attribute the language priors to the confusion of VQA model on such instances. To solve this problem, we propose a novel training framework that explicitly encourages the VQA model to distinguish between the superficially similar instances. Specifically, for each training instance, we first construct a set that contains its superficially similar counterparts. Then we exploit the proposed distinguishing module to increase the distance between the instance and its counterparts in the answer space. In this way, the VQA model is forced to further focus on the other parts of the input beyond the question type, which helps to overcome the language priors. Experimental results show that our method achieves the state-of-the-art performance on VQA-CP v2. Codes are available at Distinguishing-VQA.