Search

We present the TRIAGE Benchmark, a novel machine ethics (ME) benchmark that tests LLMs' ability to make ethical decisions during mass casualty incidents. It uses real-world ethical dilemmas with clear solutions designed by medical professionals, offering a more realistic alternative to annotation-based benchmarks. TRIAGE incorporates various prompting styles to evaluate model performance across different contexts. Most models consistently outperformed random guessing, suggesting LLMs may support decision-making in triage scenarios. Neutral or factual scenario formulations led to the best performance, unlike other ME benchmarks where ethical reminders improved outcomes. Adversarial prompts reduced performance but not to random guessing levels. Open-source models made more morally serious errors, and general capability overall predicted better performance.

In this review paper, we delve into the realm of Large Language Models (LLMs), covering their foundational principles, diverse applications, and nuanced training processes. The article sheds light on the mechanics of in-context learning and a spectrum of fine-tuning approaches, with a special focus on methods that optimize efficiency in parameter usage. Additionally, it explores how LLMs can be more closely aligned with human preferences through innovative reinforcement learning frameworks and other novel methods that incorporate human feedback. The article also examines the emerging technique of retrieval augmented generation, integrating external knowledge into LLMs. The ethical dimensions of LLM deployment are discussed, underscoring the need for mindful and responsible application. Concluding with a perspective on future research trajectories, this review offers a succinct yet comprehensive overview of the current state and emerging trends in the evolving landscape of LLMs, serving as an insightful guide for both researchers and practitioners in artificial intelligence.

Large language models (LLMs) demonstrate outstanding capabilities, but challenges remain regarding their ability to solve complex reasoning tasks, as well as their transparency, robustness, truthfulness, and ethical alignment. In this preliminary study, we compile a set of core principles for steering and evaluating the reasoning of LLMs by curating literature from several relevant strands of work: structured reasoning in LLMs, self-evaluation/self-reflection, explainability, AI system safety/security, guidelines for human critical thinking, and ethical/regulatory guidelines for AI. We identify and curate a list of 220 principles from literature, and derive a set of 37 core principles organized into seven categories: assumptions and perspectives, reasoning, information and evidence, robustness and security, ethics, utility, and implications. We conduct a small-scale expert survey, eliciting the subjective importance experts assign to different principles and lay out avenues for future work beyond our preliminary results. We envision that the development of a shared model of principles can serve multiple purposes: monitoring and steering models at inference time, improving model behavior during training, and guiding human evaluation of model reasoning., Comment: Accepted at Socially Responsible Language Modelling Research (SoLaR) workshop, NeurIPS 2023 (https://openreview.net/forum?id=8iXdNXW34d). Based on previous manuscript version: doi:10.2139/ssrn.4446991

Systematic literature reviews (SLRs) play an essential role in summarising, synthesising and validating scientific evidence. In recent years, there has been a growing interest in using machine learning techniques to automate the identification of relevant studies for SLRs. However, the lack of standardised evaluation datasets makes comparing the performance of such automated literature screening systems difficult. In this paper, we analyse the citation screening evaluation datasets, revealing that many of the available datasets are either too small, suffer from data leakage or have limited applicability to systems treating automated literature screening as a classification task, as opposed to, for example, a retrieval or question-answering task. To address these challenges, we introduce CSMeD, a meta-dataset consolidating nine publicly released collections, providing unified access to 325 SLRs from the fields of medicine and computer science. CSMeD serves as a comprehensive resource for training and evaluating the performance of automated citation screening models. Additionally, we introduce CSMeD-FT, a new dataset designed explicitly for evaluating the full text publication screening task. To demonstrate the utility of CSMeD, we conduct experiments and establish baselines on new datasets., Comment: Accepted at NeurIPS 2023 Datasets and Benchmarks Track

Emergent chain-of-thought (CoT) reasoning capabilities promise to improve performance and explainability of large language models (LLMs). However, uncertainties remain about how reasoning strategies formulated for previous model generations generalize to new model generations and different datasets. In this small-scale study, we compare different reasoning strategies induced by zero-shot prompting across six recently released LLMs (davinci-002, davinci-003, GPT-3.5-turbo, GPT-4, Flan-T5-xxl and Cohere command-xlarge) on a mixture of six question-answering datasets, including datasets from scientific and medical domains. Our findings demonstrate that while some variations in effectiveness occur, gains from CoT reasoning strategies remain robust across different models and datasets. GPT-4 has the most benefit from current state-of-the-art reasoning strategies and exhibits the best performance by applying a prompt previously discovered through automated discovery.

In recent years, deep neural networks have been widely used for building high-performance Artificial Intelligence (AI) systems for computer vision applications. Object detection is a fundamental task in computer vision, which has been greatly progressed through developing large and intricate AI models. However, the lack of transparency is a big challenge that may not allow the widespread adoption of these models. Explainable artificial intelligence is a field of research where methods are developed to help users understand the behavior, decision logics, and vulnerabilities of AI systems. Previously, few explanation methods were developed for object detection based on random masking. However, random masks may raise some issues regarding the actual importance of pixels within an image. In this paper, we design and implement a black-box explanation method named Black-box Object Detection Explanation by Masking (BODEM) through adopting a hierarchical random masking approach for object detection systems. We propose a hierarchical random masking framework in which coarse-grained masks are used in lower levels to find salient regions within an image, and fine-grained mask are used to refine the salient regions in higher levels. Experimentations on various object detection datasets and models showed that BODEM can effectively explain the behavior of object detectors. Moreover, our method outperformed Detector Randomized Input Sampling for Explanation (D-RISE) and Local Interpretable Model-agnostic Explanations (LIME) with respect to different quantitative measures of explanation effectiveness. The experimental results demonstrate that BODEM can be an effective method for explaining and validating object detection systems in black-box testing scenarios.

Clinical notes containing valuable patient information are written by different health care providers with various scientific levels and writing styles. It might be helpful for clinicians and researchers to understand what information is essential when dealing with extensive electronic medical records. Entities recognizing and mapping them to standard terminologies is crucial in reducing ambiguity in processing clinical notes. Although named entity recognition and entity linking are critical steps in clinical natural language processing, they can also result in the production of repetitive and low-value concepts. In other hand, all parts of a clinical text do not share the same importance or content in predicting the patient's condition. As a result, it is necessary to identify the section in which each content is recorded and also to identify key concepts to extract meaning from clinical texts. In this study, these challenges have been addressed by using clinical natural language processing techniques. In addition, in order to identify key concepts, a set of popular unsupervised key phrase extraction methods has been verified and evaluated. Considering that most of the clinical concepts are in the form of multi-word expressions and their accurate identification requires the user to specify n-gram range, we have proposed a shortcut method to preserve the structure of the expression based on TF-IDF. In order to evaluate the pre-processing method and select the concepts, we have designed two types of downstream tasks (multiple and binary classification) using the capabilities of transformer-based models. The obtained results show the superiority of proposed method in combination with SciBERT model, also offer an insight into the efficacy of general extracting essential phrase methods for clinical notes.

Large language models (LLMs) such as GPT-4 have recently demonstrated impressive results across a wide range of tasks. LLMs are still limited, however, in that they frequently fail at complex reasoning, their reasoning processes are opaque, they are prone to 'hallucinate' facts, and there are concerns about their underlying biases. Letting models verbalize reasoning steps as natural language, a technique known as chain-of-thought prompting, has recently been proposed as a way to address some of these issues. Here we present ThoughtSource, a meta-dataset and software library for chain-of-thought (CoT) reasoning. The goal of ThoughtSource is to improve future artificial intelligence systems by facilitating qualitative understanding of CoTs, enabling empirical evaluations, and providing training data. This first release of ThoughtSource integrates seven scientific/medical, three general-domain and five math word question answering datasets., Comment: Revision: added datasets, formatting

Large language models (LLMs) have been shown to be able to perform new tasks based on a few demonstrations or natural language instructions. While these capabilities have led to widespread adoption, most LLMs are developed by resource-rich organizations and are frequently kept from the public. As a step towards democratizing this powerful technology, we present BLOOM, a 176B-parameter open-access language model designed and built thanks to a collaboration of hundreds of researchers. BLOOM is a decoder-only Transformer language model that was trained on the ROOTS corpus, a dataset comprising hundreds of sources in 46 natural and 13 programming languages (59 in total). We find that BLOOM achieves competitive performance on a wide variety of benchmarks, with stronger results after undergoing multitask prompted finetuning. To facilitate future research and applications using LLMs, we publicly release our models and code under the Responsible AI License.

Training and evaluating language models increasingly requires the construction of meta-datasets --diverse collections of curated data with clear provenance. Natural language prompting has recently lead to improved zero-shot generalization by transforming existing, supervised datasets into a diversity of novel pretraining tasks, highlighting the benefits of meta-dataset curation. While successful in general-domain text, translating these data-centric approaches to biomedical language modeling remains challenging, as labeled biomedical datasets are significantly underrepresented in popular data hubs. To address this challenge, we introduce BigBIO a community library of 126+ biomedical NLP datasets, currently covering 12 task categories and 10+ languages. BigBIO facilitates reproducible meta-dataset curation via programmatic access to datasets and their metadata, and is compatible with current platforms for prompt engineering and end-to-end few/zero shot language model evaluation. We discuss our process for task schema harmonization, data auditing, contribution guidelines, and outline two illustrative use cases: zero-shot evaluation of biomedical prompts and large-scale, multi-task learning. BigBIO is an ongoing community effort and is available at https://github.com/bigscience-workshop/biomedical, Comment: Submitted to NeurIPS 2022 Datasets and Benchmarks Track

Measuring the performance of natural language processing models is challenging. Traditionally used metrics, such as BLEU and ROUGE, originally devised for machine translation and summarization, have been shown to suffer from low correlation with human judgment and a lack of transferability to other tasks and languages. In the past 15 years, a wide range of alternative metrics have been proposed. However, it is unclear to what extent this has had an impact on NLP benchmarking efforts. Here we provide the first large-scale cross-sectional analysis of metrics used for measuring performance in natural language processing. We curated, mapped and systematized more than 3500 machine learning model performance results from the open repository 'Papers with Code' to enable a global and comprehensive analysis. Our results suggest that the large majority of natural language processing metrics currently used have properties that may result in an inadequate reflection of a models' performance. Furthermore, we found that ambiguities and inconsistencies in the reporting of metrics may lead to difficulties in interpreting and comparing model performances, impairing transparency and reproducibility in NLP research., Comment: "NLP Power" workshop at ACL 2022. This work is based on a previous arXiv submission: arXiv:2008.02577 [cs.AI]

Benchmarks are crucial to measuring and steering progress in artificial intelligence (AI). However, recent studies raised concerns over the state of AI benchmarking, reporting issues such as benchmark overfitting, benchmark saturation and increasing centralization of benchmark dataset creation. To facilitate monitoring of the health of the AI benchmarking ecosystem, we introduce methodologies for creating condensed maps of the global dynamics of benchmark creation and saturation. We curated data for 3765 benchmarks covering the entire domains of computer vision and natural language processing, and show that a large fraction of benchmarks quickly trended towards near-saturation, that many benchmarks fail to find widespread utilization, and that benchmark performance gains for different AI tasks were prone to unforeseen bursts. We analyze attributes associated with benchmark popularity, and conclude that future benchmarks should emphasize versatility, breadth and real-world utility., Comment: This version includes more recent data and additional analyses

In this article, we first give an introduction to artificial intelligence and its applications in biology and medicine in Section 1. Deep learning methods are then described in Section 2. We narrow down the focus of the study on textual data in Section 3, where natural language processing and its applications in the biomedical domain are described. In Section 4, we give an introduction to explainable artificial intelligence and discuss the importance of explainability of artificial intelligence systems, especially in the biomedical domain.

Publicly accessible benchmarks that allow for assessing and comparing model performances are important drivers of progress in artificial intelligence (AI). While recent advances in AI capabilities hold the potential to transform medical practice by assisting and augmenting the cognitive processes of healthcare professionals, the coverage of clinically relevant tasks by AI benchmarks is largely unclear. Furthermore, there is a lack of systematized meta-information that allows clinical AI researchers to quickly determine accessibility, scope, content and other characteristics of datasets and benchmark datasets relevant to the clinical domain. To address these issues, we curated and released a comprehensive catalogue of datasets and benchmarks pertaining to the broad domain of clinical and biomedical natural language processing (NLP), based on a systematic review of literature and online resources. A total of 450 NLP datasets were manually systematized and annotated with rich metadata, such as targeted tasks, clinical applicability, data types, performance metrics, accessibility and licensing information, and availability of data splits. We then compared tasks covered by AI benchmark datasets with relevant tasks that medical practitioners reported as highly desirable targets for automation in a previous empirical study. Our analysis indicates that AI benchmarks of direct clinical relevance are scarce and fail to cover most work activities that clinicians want to see addressed. In particular, tasks associated with routine documentation and patient data administration workflows are not represented despite significant associated workloads. Thus, currently available AI benchmarks are improperly aligned with desired targets for AI automation in clinical settings, and novel benchmarks should be created to fill these gaps., Comment: (this version extends the literature references)

Deep transformer neural network models have improved the predictive accuracy of intelligent text processing systems in the biomedical domain. They have obtained state-of-the-art performance scores on a wide variety of biomedical and clinical Natural Language Processing (NLP) benchmarks. However, the robustness and reliability of these models has been less explored so far. Neural NLP models can be easily fooled by adversarial samples, i.e. minor changes to input that preserve the meaning and understandability of the text but force the NLP system to make erroneous decisions. This raises serious concerns about the security and trust-worthiness of biomedical NLP systems, especially when they are intended to be deployed in real-world use cases. We investigated the robustness of several transformer neural language models, i.e. BioBERT, SciBERT, BioMed-RoBERTa, and Bio-ClinicalBERT, on a wide range of biomedical and clinical text processing tasks. We implemented various adversarial attack methods to test the NLP systems in different attack scenarios. Experimental results showed that the biomedical NLP models are sensitive to adversarial samples; their performance dropped in average by 21 and 18.9 absolute percent on character-level and word-level adversarial noise, respectively. Conducting extensive adversarial training experiments, we fine-tuned the NLP models on a mixture of clean samples and adversarial inputs. Results showed that adversarial training is an effective defense mechanism against adversarial noise; the models robustness improved in average by 11.3 absolute percent. In addition, the models performance on clean data increased in average by 2.4 absolute present, demonstrating that adversarial training can boost generalization abilities of biomedical NLP systems.