Recent advances in large language models (LLMs) have led to significant improvements in translating natural language questions into SQL queries. While achieving high accuracy in SQL generation is crucial, little is known about the extent to which these text-to-SQL models can reliably handle diverse types of questions encountered during real-world deployment, including unanswerable ones. To explore this aspect, we present TrustSQL, a new benchmark designed to assess the reliability of text-to-SQL models in both single-database and cross-database settings. The benchmark tasks models with providing one of two outcomes: 1) SQL prediction; or 2) abstention from making a prediction, either when there is a potential error in the generated SQL or when faced with unanswerable questions. For model evaluation, we explore various modeling approaches specifically designed for this task. These include: 1) optimizing separate models for answerability detection, SQL generation, and error detection, which are then integrated into a single pipeline; and 2) developing a unified approach that optimizes a single model to address the proposed task. Experimental results using our new reliability score show that addressing this challenge involves many different areas of research and opens new avenues for model development. Nonetheless, none of the methods surpass the reliability performance of the naive baseline, which abstains from answering all questions.
We introduce KorMedMCQA, the first Korean multiple-choice question answering (MCQA) benchmark derived from Korean healthcare professional licensing examinations, covering from the year 2012 to year 2023. This dataset consists of a selection of questions from the license examinations for doctors, nurses, and pharmacists, featuring a diverse array of subjects. We conduct baseline experiments on various large language models, including proprietary/open-source, multilingual/Korean-additional pretrained, and clinical context pretrained models, highlighting the potential for further enhancements. We make our data publicly available on HuggingFace (https://huggingface.co/datasets/sean0042/KorMedMCQA) and provide a evaluation script via LM-Harness, inviting further exploration and advancement in Korean healthcare environments.
The third ML4H symposium was held in person on December 10, 2023, in New Orleans, Louisiana, USA. The symposium included research roundtable sessions to foster discussions between participants and senior researchers on timely and relevant topics for the \ac{ML4H} community. Encouraged by the successful virtual roundtables in the previous year, we organized eleven in-person roundtables and four virtual roundtables at ML4H 2022. The organization of the research roundtables at the conference involved 17 Senior Chairs and 19 Junior Chairs across 11 tables. Each roundtable session included invited senior chairs (with substantial experience in the field), junior chairs (responsible for facilitating the discussion), and attendees from diverse backgrounds with interest in the session's topic. Herein we detail the organization process and compile takeaways from these roundtable discussions, including recent advances, applications, and open challenges for each topic. We conclude with a summary and lessons learned across all roundtables. This document serves as a comprehensive review paper, summarizing the recent advancements in machine learning for healthcare as contributed by foremost researchers in the field.
This study introduces EHRNoteQA, a novel patient-specific question answering benchmark tailored for evaluating Large Language Models (LLMs) in clinical environments. Based on MIMIC-IV Electronic Health Record (EHR), a team of three medical professionals has curated the dataset comprising 962 unique questions, each linked to a specific patient's EHR clinical notes. What makes EHRNoteQA distinct from existing EHR-based benchmarks is as follows: Firstly, it is the first dataset to adopt a multi-choice question answering format, a design choice that effectively evaluates LLMs with reliable scores in the context of automatic evaluation, compared to other formats. Secondly, it requires an analysis of multiple clinical notes to answer a single question, reflecting the complex nature of real-world clinical decision-making where clinicians review extensive records of patient histories. Our comprehensive evaluation on various large language models showed that their scores on EHRNoteQA correlate more closely with their performance in addressing real-world medical questions evaluated by clinicians than their scores from other LLM benchmarks. This underscores the significance of EHRNoteQA in evaluating LLMs for medical applications and highlights its crucial role in facilitating the integration of LLMs into healthcare systems. The dataset will be made available to the public under PhysioNet credential access, promoting further research in this vital field.
Transformer-based models have significantly improved performance across a range of multimodal understanding tasks, such as visual question answering and action recognition. However, multimodal Transformers significantly suffer from a quadratic complexity of the multi-head attention with the input sequence length, especially as the number of modalities increases. To address this, we introduce Low-Cost Multimodal Transformer (LoCoMT), a novel multimodal attention mechanism that aims to reduce computational cost during training and inference with minimal performance loss. Specifically, by assigning different multimodal attention patterns to each attention head, LoCoMT can flexibly control multimodal signals and theoretically ensures a reduced computational cost compared to existing multimodal Transformer variants. Experimental results on two multimodal datasets, namely Audioset and MedVidCL demonstrate that LoCoMT not only reduces GFLOPs but also matches or even outperforms established models.
For Large Language Models (LLMs) to be effectively deployed in a specific country, they must possess an understanding of the nation's culture and basic knowledge. To this end, we introduce National Alignment, which measures an alignment between an LLM and a targeted country from two aspects: social value alignment and common knowledge alignment. Social value alignment evaluates how well the model understands nation-specific social values, while common knowledge alignment examines how well the model captures basic knowledge related to the nation. We constructed KorNAT, the first benchmark that measures national alignment with South Korea. For the social value dataset, we obtained ground truth labels from a large-scale survey involving 6,174 unique Korean participants. For the common knowledge dataset, we constructed samples based on Korean textbooks and GED reference materials. KorNAT contains 4K and 6K multiple-choice questions for social value and common knowledge, respectively. Our dataset creation process is meticulously designed and based on statistical sampling theory and was refined through multiple rounds of human review. The experiment results of seven LLMs reveal that only a few models met our reference score, indicating a potential for further enhancement. KorNAT has received government approval after passing an assessment conducted by a government-affiliated organization dedicated to evaluating dataset quality. Samples and detailed evaluation protocols of our dataset can be found in https://selectstar.ai/ko/papers-national-alignment
Long-term forecasting presents unique challenges due to the time and memory complexity of handling long sequences. Existing methods, which rely on sliding windows to process long sequences, struggle to effectively capture long-term variations that are partially caught within the short window (i.e., outer-window variations). In this paper, we introduce a novel approach that overcomes this limitation by employing contrastive learning and enhanced decomposition architecture, specifically designed to focus on long-term variations. To this end, our contrastive loss incorporates global autocorrelation held in the whole time series, which facilitates the construction of positive and negative pairs in a self-supervised manner. When combined with our decomposition networks, our contrastive learning significantly improves long-term forecasting performance. Extensive experiments demonstrate that our approach outperforms 14 baseline models in multiple experiments over nine long-term benchmarks, especially in challenging scenarios that require a significantly long output for forecasting. Source code is available at https://github.com/junwoopark92/Self-Supervised-Contrastive-Forecsating.
Wearable technologies enable continuous monitoring of various health metrics, such as physical activity, heart rate, sleep, and stress levels. A key challenge with wearable data is obtaining quality labels. Unlike modalities like video where the videos themselves can be effectively used to label objects or events, wearable data do not contain obvious cues about the physical manifestation of the users and usually require rich metadata. As a result, label noise can become an increasingly thorny issue when labeling such data. In this paper, we propose a novel solution to address noisy label learning, entitled Few-Shot Human-in-the-Loop Refinement (FHLR). Our method initially learns a seed model using weak labels. Next, it fine-tunes the seed model using a handful of expert corrections. Finally, it achieves better generalizability and robustness by merging the seed and fine-tuned models via weighted parameter averaging. We evaluate our approach on four challenging tasks and datasets, and compare it against eight competitive baselines designed to deal with noisy labels. We show that FHLR achieves significantly better performance when learning from noisy labels and achieves state-of-the-art by a large margin, with up to 19% accuracy improvement under symmetric and asymmetric noise. Notably, we find that FHLR is particularly robust to increased label noise, unlike prior works that suffer from severe performance degradation. Our work not only achieves better generalization in high-stakes health sensing benchmarks but also sheds light on how noise affects commonly-used models.
Developing clinical prediction models (e.g., mortality prediction) based on electronic health records (EHRs) typically relies on expert opinion for feature selection and adjusting observation window size. This burdens experts and creates a bottleneck in the development process. We propose Retrieval-Enhanced Medical prediction model (REMed) to address such challenges. REMed can essentially evaluate an unlimited number of clinical events, select the relevant ones, and make predictions. This approach effectively eliminates the need for manual feature selection and enables an unrestricted observation window. We verified these properties through experiments on 27 clinical tasks and two independent cohorts from publicly available EHR datasets, where REMed outperformed other contemporary architectures that aim to handle as many events as possible. Notably, we found that the preferences of REMed align closely with those of medical experts. We expect our approach to significantly expedite the development of EHR prediction models by minimizing clinicians' need for manual involvement.