We study the behavior of several black-box search algorithms used for generating adversarial examples for natural language processing (NLP) tasks. We perform a fine-grained analysis of three elements relevant to search: search algorithm, search space, and search budget. When new search algorithms are proposed in past work, the attack search space is often modified alongside the search algorithm. Without ablation studies benchmarking the search algorithm change with the search space held constant, one cannot tell if an increase in attack success rate is a result of an improved search algorithm or a less restrictive search space. Additionally, many previous studies fail to properly consider the search algorithms' run-time cost, which is essential for downstream tasks like adversarial training. Our experiments provide a reproducible benchmark of search algorithms across a variety of search spaces and query budgets to guide future research in adversarial NLP. Based on our experiments, we recommend greedy attacks with word importance ranking when under a time constraint or attacking long inputs, and either beam search or particle swarm optimization otherwise. Code implementation shared via https://github.com/QData/TextAttack-Search-Benchmark
Facebook AI Research（脸书人工智能研究院）；UCL（伦敦大学学院）；UNC Chapel Hill（北卡罗来纳大学教堂山分校）；CMU（卡内基梅隆大学）；Stanford University（斯坦福大学）；Alan Turing Institute（艾伦·图灵研究所）；JHU（约翰斯·霍普金斯大学）；Simon Fraser University（西蒙弗雷泽大学）
We introduce Dynabench, an open-source platform for dynamic dataset creation and model benchmarking. Dynabench runs in a web browser and supports human-and-model-in-the-loop dataset creation: annotators seek to create examples that a target model will misclassify, but that another person will not. In this paper, we argue that Dynabench addresses a critical need in our community: contemporary models quickly achieve outstanding performance on benchmark tasks but nonetheless fail on simple challenge examples and falter in real-world scenarios. With Dynabench, dataset creation, model development, …
University of Science and Technology of China（中国科学技术大学）；UNSW Sydney（新南威尔士大学）；Genmab Inc.（Genmab公司）
NLP-based computer vision models, particularly vision transformers, have been shown to outperform CNN models in many imaging tasks. However, most digital pathology artificial-intelligence models are based on CNN architectures, probably owing to a lack of data regarding NLP models for pathology images. In this study, we developed digital pathology pipelines to benchmark the five most recently proposed NLP models (vision transformer (ViT), Swin Transformer, MobileViT, CMT, and Sequencer2D) and four popular CNN models (ResNet18, ResNet50, MobileNetV2, and EfficientNet) to predict biomarkers in co…
University of Southern California（南加州大学）；Amazon Alexa AI（亚马逊Alexa AI）
Increasing concerns and regulations about data privacy and sparsity necessitate the study of privacy-preserving, decentralized learning methods for natural language processing (NLP) tasks. Federated learning (FL) provides promising approaches for a large number of clients (e.g., personal devices or organizations) to collaboratively learn a shared global model to benefit all clients while allowing users to keep their data locally. Despite interest in studying FL methods for NLP tasks, a systematic comparison and analysis is lacking in the literature. Herein, we present the FedNLP, a benchmarkin…
In Machine Learning, a benchmark refers to an ensemble of datasets associated with one or multiple metrics together with a way to aggregate different systems performances. They are instrumental in (i) assessing the progress of new methods along different axes and (ii) selecting the best systems for practical use. This is particularly the case for NLP with the development of large pre-trained models (e.g. GPT, BERT) that are expected to generalize well on a variety of tasks. While the community mainly focused on developing new datasets and metrics, there has been little interest in the aggregat…
We introduce Dynaboard, an evaluation-as-a-service framework for hosting benchmarks and conducting holistic model comparison, integrated with the Dynabench platform. Our platform evaluates NLP models directly instead of relying on self-reported metrics or predictions on a single dataset. Under this paradigm, models are submitted to be evaluated in the cloud, circumventing the issues of reproducibility, accessibility, and backwards compatibility that often hinder benchmarking in NLP. This allows users to interact with uploaded models in real time to assess their quality, and permits the collect…
School of Computer Science, Fudan University（复旦大学计算机学院）；Shanghai Key Laboratory of Intelligent Information Processing（上海市智能信息处理重点实验室）；Department of Computer Science, University of California, Los Angeles（加利福尼亚大学洛杉矶分校计算机科学系）
Recent studies have shown that deep neural networks are vulnerable to intentionally crafted adversarial examples, and various methods have been proposed to defend against adversarial word-substitution attacks for neural NLP models. However, there is a lack of systematic study on comparing different defense approaches under the same attacking setting. In this paper, we seek to fill the gap of systematic studies through comprehensive researches on understanding the behavior of neural text classifiers trained by various defense methods under representative adversarial attacks. In addition, we pro…
Although recent Massively Multilingual Language Models (MMLMs) like mBERT and XLMR support around 100 languages, most existing multilingual NLP benchmarks provide evaluation data in only a handful of these languages with little linguistic diversity. We argue that this makes the existing practices in multilingual evaluation unreliable and does not provide a full picture of the performance of MMLMs across the linguistic landscape. We propose that the recent work done in Performance Prediction for NLP tasks can serve as a potential solution in fixing benchmarking in Multilingual NLP by utilizing …
Benchmarking the Performance of Pre-trained LLMs across Urdu NLP Tasks
Center for Language Engineering（语言工程中心）；Al-Khawarizmi Institute of Computer Science（阿尔-花剌子密计算机科学研究所）；University of Engineering and Technology Lahore（拉合尔工程技术大学）
Manipal Institute of Technology（马尼帕尔理工学院）；Carnegie Mellon University（卡内基梅隆大学）；OpenMined；Yildiz Technical University（耶尔德兹技术大学）；Ford Motor Company（福特汽车公司）；University of California, San Diego（加州大学圣地亚哥分校）
Natural Language Processing (NLP) techniques can be applied to help with the diagnosis of medical conditions such as depression, using a collection of a person's utterances. Depression is a serious medical illness that can have adverse effects on how one feels, thinks, and acts, which can lead to emotional and physical problems. Due to the sensitive nature of such data, privacy measures need to be taken for handling and training models with such data. In this work, we study the effects that the application of Differential Privacy (DP) has, in both a centralized and a Federated Learning (FL) se…
One Law, Many Languages: Benchmarking Multilingual Legal Reasoning for Judicial Support
University of Bern（伯尔尼大学）；University of Copenhagen（哥本哈根大学）；Bern University of Applied Sciences（伯尔尼应用科学大学）；University of Fribourg（弗里堡大学）；Stanford University（斯坦福大学）
Wroclaw University of Science and Technology（弗罗茨瓦夫理工大学）；Poznan University of Technology（波兹南理工大学）
The availability of compute and data to train larger and larger language models increases the demand for robust methods of benchmarking the true progress of LM training. Recent years witnessed significant progress in standardized benchmarking for English. Benchmarks such as GLUE, SuperGLUE, or KILT have become de facto standard tools to compare large language models. Following the trend to replicate GLUE for other languages, the KLEJ benchmark has been released for Polish. In this paper, we evaluate the progress in benchmarking for low-resourced languages. We note that only a handful of langua…
UniIR: Training and Benchmarking Universal Multimodal Information Retrievers
University of Waterloo（滑铁卢大学）；Georgia Institute of Technology（佐治亚理工学院）；Hong Kong University of Science and Technology（香港科技大学）；Google DeepMind（谷歌DeepMind）
AD-LLM: Benchmarking Large Language Models for Anomaly Detection
University of Southern California（南加州大学）；Northwestern University（西北大学）；Arizona State University（亚利桑那州立大学）；Adobe Research（Adobe研究）；Rice University（里士满大学）
Bosch Center for Artificial Intelligence（博世人工智能中心）；Center for Information and Language Processing, LMU Munich（慕尼黑大学信息与语言处理中心）
Many natural language processing (NLP) tasks are naturally imbalanced, as some target categories occur much more frequently than others in the real world. In such scenarios, current NLP models still tend to perform poorly on less frequent classes. Addressing class imbalance in NLP is an active research topic, yet, finding a good approach for a particular task and imbalance scenario is difficult. With this survey, the first overview on class imbalance in deep-learning based NLP, we provide guidance for NLP researchers and practitioners dealing with imbalanced data. We first discuss various type…
In recent times, there has been definitive progress in the field of NLP, with its applications growing as the utility of our language models increases with advances in their performance. However, these models require a large amount of computational power and data to train, consequently leading to large carbon footprints. Therefore, it is imperative that we study the carbon efficiency and look for alternatives to reduce the overall environmental impact of training models, in particular large language models. In our work, we assess the performance of models for machine translation, across multip…
National Center for Biotechnology Information（国家生物技术信息中心）；National Library of Medicine, National Institutes of Health（美国国立卫生研究院国家医学图书馆）
Inspired by the success of the General Language Understanding Evaluation benchmark, we introduce the Biomedical Language Understanding Evaluation (BLUE) benchmark to facilitate research in the development of pre-training language representations in the biomedicine domain. The benchmark consists of five tasks with ten datasets that cover both biomedical and clinical texts with different dataset sizes and difficulties. We also evaluate several baselines based on BERT and ELMo and find that the BERT model pre-trained on PubMed abstracts and MIMIC-III clinical notes achieves the best results. We m…
Consolidating and Developing Benchmarking Datasets for the Nepali Natural Language Understanding Tasks
Information and Language Processing Research Lab (ILPRL)（信息与语言处理研究实验室）；Kathmandu University（加德满都大学）；Institute for Research and Innovation in Intelligent Systems (IRIIS)（智能系统研究与创新研究院）
Benchmarking Gender and Political Bias in Large Language Models
School of Computing & Information Systems, The University of Melbourne（墨尔本大学计算与信息系统学院）；Mohamed bin Zayed University of Artificial Intelligence（穆罕默德·本·扎耶德人工智能大学）
Validate Your Authority: Benchmarking LLMs on Multi-Label Precedent Treatment Classification
Department of Information Science and Technology（信息科学与技术系）；College of Emergency Preparedness, Homeland Security, and Cybersecurity（应急准备、国土安全与网络安全学院）；University at Albany, SUNY（萨利纳大学）
Reasoning over commonsense knowledge bases (CSKB) whose elements are in the form of free-text is an important yet hard task in NLP. While CSKB completion only fills the missing links within the domain of the CSKB, CSKB population is alternatively proposed with the goal of reasoning unseen assertions from external resources. In this task, CSKBs are grounded to a large-scale eventuality (activity, state, and event) graph to discriminate whether novel triples from the eventuality graph are plausible or not. However, existing evaluations on the population task are either not accurate (automatic ev…
