Exploring Large Language Models in Healthcare: Insights into Corpora Sources, Customization Strategies, and Evaluation Metrics

This study reviewed the use of Large Language Models (LLMs) in healthcare, focusing on their training corpora, customization techniques, and evaluation metrics. A systematic search of studies from 2021 to 2024 identified 61 articles. Four types of corpora were used: clinical resources, literature, open-source datasets, and web-crawled data. Common construction techniques included pre-training, prompt engineering, and retrieval-augmented generation, with 44 studies combining multiple methods. Evaluation metrics were categorized into process, usability, and outcome metrics, with outcome metrics divided into model-based and expert-assessed outcomes. The study identified critical gaps in corpus fairness, which contributed to biases from geographic, cultural, and socio-economic factors. The reliance on unverified or unstructured data highlighted the need for better integration of evidence-based clinical guidelines. Future research should focus on developing a tiered corpus architecture with vetted sources and dynamic weighting, while ensuring model transparency. Additionally, the lack of standardized evaluation frameworks for domain-specific models called for comprehensive validation of LLMs in real-world healthcare settings.

Towards Understanding Fine-Tuning Mechanisms of LLMs via Circuit Analysis

Fine-tuning significantly improves the performance of Large Language Models (LLMs), yet its underlying mechanisms remain poorly understood. This paper aims to provide an in-depth interpretation of the fine-tuning process through circuit analysis, a popular tool in Mechanistic Interpretability (MI). Unlike previous studies \cite{prakash2024finetuningenhancesexistingmechanisms,chhabra2024neuroplasticity} that focus on tasks where pre-trained models already perform well, we develop a set of mathematical tasks where fine-tuning yields substantial performance gains, which are closer to the practical setting. In our experiments, we identify circuits at various checkpoints during fine-tuning and examine the interplay between circuit analysis, fine-tuning methods, and task complexities. First, we find that while circuits maintain high node similarity before and after fine-tuning, their edges undergo significant changes, which is in contrast to the previous work \cite{prakash2024finetuningenhancesexistingmechanisms,chhabra2024neuroplasticity} that show circuits only add some additional components after fine-tuning. Based on these observations, we develop a circuit-aware Low-Rank Adaptation (LoRA) method, which assigns ranks to layers based on edge changes in the circuits. Experimental results demonstrate that our circuit-based LoRA algorithm achieves an average performance improvement of 2.46\% over standard LoRA with similar parameter sizes. Furthermore, we explore how combining circuits from subtasks can enhance fine-tuning in compositional tasks, providing new insights into the design of such tasks and deepening the understanding of circuit dynamics and fine-tuning mechanisms.

Federated Linear Dueling Bandits

Contextual linear dueling bandits have recently garnered significant attention due to their widespread applications in important domains such as recommender systems and large language models. Classical dueling bandit algorithms are typically only applicable to a single agent. However, many applications of dueling bandits involve multiple agents who wish to collaborate for improved performance yet are unwilling to share their data. This motivates us to draw inspirations from federated learning, which involves multiple agents aiming to collaboratively train their neural networks via gradient descent (GD) without sharing their raw data. Previous works have developed federated linear bandit algorithms which rely on closed-form updates of the bandit parameters (e.g., the linear function parameter) to achieve collaboration. However, in linear dueling bandits, the linear function parameter lacks a closed-form expression and its estimation requires minimizing a loss function. This renders these previous methods inapplicable. In this work, we overcome this challenge through an innovative and principled combination of online gradient descent (for minimizing the loss function to estimate the linear function parameters) and federated learning, hence introducing the first federated linear dueling bandit algorithms. Through rigorous theoretical analysis, we prove that our algorithms enjoy a sub-linear upper bound on its cumulative regret. We also use empirical experiments to demonstrate the effectiveness of our algorithms and the practical benefit of collaboration.

COph100: A comprehensive fundus image registration dataset from infants constituting the "RIDIRP" database

Retinal image registration is vital for diagnostic therapeutic applications within the field of ophthalmology. Existing public datasets, focusing on adult retinal pathologies with high-quality images, have limited number of image pairs and neglect clinical challenges. To address this gap, we introduce COph100, a novel and challenging dataset known as the Comprehensive Ophthalmology Retinal Image Registration dataset for infants with a wide range of image quality issues constituting the public "RIDIRP" database. COph100 consists of 100 eyes, each with 2 to 9 examination sessions, amounting to a total of 491 image pairs carefully selected from the publicly available dataset. We manually labeled the corresponding ground truth image points and provided automatic vessel segmentation masks for each image. We have assessed COph100 in terms of image quality and registration outcomes using state-of-the-art algorithms. This resource enables a robust comparison of retinal registration methodologies and aids in the analysis of disease progression in infants, thereby deepening our understanding of pediatric ophthalmic conditions.

BlenderLLM: Training Large Language Models for Computer-Aided Design with Self-improvement

The application of Large Language Models (LLMs) in Computer-Aided Design (CAD) remains an underexplored area, despite their remarkable advancements in other domains. In this paper, we present BlenderLLM, a novel framework for training LLMs specifically for CAD tasks leveraging a self-improvement methodology. To support this, we developed a bespoke training dataset, BlendNet, and introduced a comprehensive evaluation suite, CADBench. Our results reveal that existing models demonstrate significant limitations in generating accurate CAD scripts. However, through minimal instruction-based fine-tuning and iterative self-improvement, BlenderLLM significantly surpasses these models in both functionality and accuracy of CAD script generation. This research establishes a strong foundation for the application of LLMs in CAD while demonstrating the transformative potential of self-improving models in advancing CAD automation. We encourage further exploration and adoption of these methodologies to drive innovation in the field. The dataset, model, benchmark, and source code are publicly available at https://github.com/FreedomIntelligence/BlenderLLM

CDEMapper: Enhancing NIH Common Data Element Normalization using Large Language Models

Common Data Elements (CDEs) standardize data collection and sharing across studies, enhancing data interoperability and improving research reproducibility. However, implementing CDEs presents challenges due to the broad range and variety of data elements. This study aims to develop an effective and efficient mapping tool to bridge the gap between local data elements and National Institutes of Health (NIH) CDEs. We propose CDEMapper, a large language model (LLM) powered mapping tool designed to assist in mapping local data elements to NIH CDEs. CDEMapper has three core modules: (1) CDE indexing and embeddings. NIH CDEs were indexed and embedded to support semantic search; (2) CDE recommendations. The tool combines Elasticsearch (BM25 similarity methods) with state of the art GPT services to recommend candidate CDEs and their permissible values; and (3) Human review. Users review and select the NIH CDEs and values that best match their data elements and value sets. We evaluate the tool recommendation accuracy against manually annotated mapping results. CDEMapper offers a publicly available, LLM-powered, and intuitive user interface that consolidates essential and advanced mapping services into a streamlined pipeline. It provides a step by step, quality assured mapping workflow designed with a user-centered approach. The evaluation results demonstrated that augmenting BM25 with GPT embeddings and a ranker consistently enhances CDEMapper mapping accuracy in three different mapping settings across four evaluation datasets. This work opens up the potential of using LLMs to assist with CDE recommendation and human curation when aligning local data elements with NIH CDEs. Additionally, this effort enhances clinical research data interoperability and helps researchers better understand the gaps between local data elements and NIH CDEs.

Memory-Augmented Multimodal LLMs for Surgical VQA via Self-Contained Inquiry

Comprehensively understanding surgical scenes in Surgical Visual Question Answering (Surgical VQA) requires reasoning over multiple objects. Previous approaches address this task using cross-modal fusion strategies to enhance reasoning ability. However, these methods often struggle with limited scene understanding and question comprehension, and some rely on external resources (e.g., pre-extracted object features), which can introduce errors and generalize poorly across diverse surgical environments. To address these challenges, we propose SCAN, a simple yet effective memory-augmented framework that leverages Multimodal LLMs to improve surgical context comprehension via Self-Contained Inquiry. SCAN operates autonomously, generating two types of memory for context augmentation: Direct Memory (DM), which provides multiple candidates (or hints) to the final answer, and Indirect Memory (IM), which consists of self-contained question-hint pairs to capture broader scene context. DM directly assists in answering the question, while IM enhances understanding of the surgical scene beyond the immediate query. Reasoning over these object-aware memories enables the model to accurately interpret images and respond to questions. Extensive experiments on three publicly available Surgical VQA datasets demonstrate that SCAN achieves state-of-the-art performance, offering improved accuracy and robustness across various surgical scenarios.

BlueLM-V-3B: Algorithm and System Co-Design for Multimodal Large Language Models on Mobile Devices

The emergence and growing popularity of multimodal large language models (MLLMs) have significant potential to enhance various aspects of daily life, from improving communication to facilitating learning and problem-solving. Mobile phones, as essential daily companions, represent the most effective and accessible deployment platform for MLLMs, enabling seamless integration into everyday tasks. However, deploying MLLMs on mobile phones presents challenges due to limitations in memory size and computational capability, making it difficult to achieve smooth and real-time processing without extensive optimization. In this paper, we present BlueLM-V-3B, an algorithm and system co-design approach specifically tailored for the efficient deployment of MLLMs on mobile platforms. To be specific, we redesign the dynamic resolution scheme adopted by mainstream MLLMs and implement system optimization for hardware-aware deployment to optimize model inference on mobile phones. BlueLM-V-3B boasts the following key highlights: (1) Small Size: BlueLM-V-3B features a language model with 2.7B parameters and a vision encoder with 400M parameters. (2) Fast Speed: BlueLM-V-3B achieves a generation speed of 24.4 token/s on the MediaTek Dimensity 9300 processor with 4-bit LLM weight quantization. (3) Strong Performance: BlueLM-V-3B has attained the highest average score of 66.1 on the OpenCompass benchmark among models with $\leq$ 4B parameters and surpassed a series of models with much larger parameter sizes (e.g., MiniCPM-V-2.6, InternVL2-8B).

Information Extraction from Clinical Notes: Are We Ready to Switch to Large Language Models?

Backgrounds: Information extraction (IE) is critical in clinical natural language processing (NLP). While large language models (LLMs) excel on generative tasks, their performance on extractive tasks remains debated. Methods: We investigated Named Entity Recognition (NER) and Relation Extraction (RE) using 1,588 clinical notes from four sources (UT Physicians, MTSamples, MIMIC-III, and i2b2). We developed an annotated corpus covering 4 clinical entities and 16 modifiers, and compared instruction-tuned LLaMA-2 and LLaMA-3 against BiomedBERT in terms of performance, generalizability, computational resources, and throughput to BiomedBERT. Results: LLaMA models outperformed BiomedBERT across datasets. With sufficient training data, LLaMA showed modest improvements (1% on NER, 1.5-3.7% on RE); improvements were larger with limited training data. On unseen i2b2 data, LLaMA-3-70B outperformed BiomedBERT by 7% (F1) on NER and 4% on RE. However, LLaMA models required more computing resources and ran up to 28 times slower. We implemented "Kiwi," a clinical IE package featuring both models, available at https://kiwi.clinicalnlp.org/. Conclusion: This study is among the first to develop and evaluate a comprehensive clinical IE system using open-source LLMs. Results indicate that LLaMA models outperform BiomedBERT for clinical NER and RE but with higher computational costs and lower throughputs. These findings highlight that choosing between LLMs and traditional deep learning methods for clinical IE applications should remain task-specific, taking into account both performance metrics and practical considerations such as available computing resources and the intended use case scenarios.

UCFE: A User-Centric Financial Expertise Benchmark for Large Language Models

This paper introduces the UCFE: User-Centric Financial Expertise benchmark, an innovative framework designed to evaluate the ability of large language models (LLMs) to handle complex real-world financial tasks. UCFE benchmark adopts a hybrid approach that combines human expert evaluations with dynamic, task-specific interactions to simulate the complexities of evolving financial scenarios. Firstly, we conducted a user study involving 804 participants, collecting their feedback on financial tasks. Secondly, based on this feedback, we created our dataset that encompasses a wide range of user intents and interactions. This dataset serves as the foundation for benchmarking 12 LLM services using the LLM-as-Judge methodology. Our results show a significant alignment between benchmark scores and human preferences, with a Pearson correlation coefficient of 0.78, confirming the effectiveness of the UCFE dataset and our evaluation approach. UCFE benchmark not only reveals the potential of LLMs in the financial sector but also provides a robust framework for assessing their performance and user satisfaction. The benchmark dataset and evaluation code are available.