paper-with-me

Papers

Trick Me If You Can: Human-in-the-loop Generation of Adversarial Examples for Question Answering

2018-09-07 · TACL 2019 3 · Eric Wallace, Pedro Rodriguez, Shi Feng, Ikuya Yamada, Jordan Boyd-Graber

Adversarial evaluation stress tests a model's understanding of natural language. While past approaches expose superficial patterns, the resulting adversarial examples are limited in complexity and diversity. We propose human-in-the-loop adversarial generation, where human authors are guided to break models. We aid the authors with interpretations of model predictions through an interactive user interface. We apply this generation framework to a question answering task called Quizbowl, where trivia enthusiasts craft adversarial questions. The resulting questions are validated via live human--computer matches: although the questions appear ordinary to humans, they systematically stump neural and information retrieval models. The adversarial questions cover diverse phenomena from multi-hop reasoning to entity type distractors, exposing open challenges in robust question answering.

📄 PDF Abstract BibTeX arXiv:1809.02701

Code (1)

Eric-Wallace/trickme-interface 공식 구현

Tasks

DiversityInformation RetrievalQuestion AnsweringRetrieval

Similar Papers 제목 키워드 기반

How the Advent of Ubiquitous Large Language Models both Stymie and Turbocharge Dynamic Adversarial Question Generation

2024-01-20 · Yoo yeon Sung, Ishani Mondal, Jordan Boyd-Graber

Dynamic adversarial question generation, where humans write examples to stump a model, aims to create examples that are realistic and informative. However, the advent of large language models (LLMs) has been a double-edg…

Question GenerationQuestion-GenerationRetrieval

Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation

2021-04-18 · EMNLP 2021 11 · Max Bartolo, Tristan Thrush, Robin Jia, Sebastian Riedel 외

Despite recent progress, state-of-the-art question answering models remain vulnerable to a variety of adversarial attacks. While dynamic adversarial data collection, in which a human annotator tries to write examples tha…

Answer SelectionQuestion AnsweringQuestion Generation

Benchmarking and Analyzing Robust Point Cloud Recognition: Bag of Tricks for Defending Adversarial Examples

2023-07-31 · ICCV 2023 1 · Qiufan Ji, Lin Wang, Cong Shi, Shengshan Hu 외

Deep Neural Networks (DNNs) for 3D point cloud recognition are vulnerable to adversarial examples, threatening their practical deployment. Despite the many research endeavors have been made to tackle this issue in recent…

Adversarial RobustnessBenchmarking

Human Adversarial QA: Did the Model Understand the Paragraph?

2020-10-16 · NeurIPS Workshop HAMLETS 2020 12 · Anonymous

Recently, adversarial attacks have become an important means of gauging the robustness of natural language models as training and testing set methodology has proved inadequate. In this paper we explore an evaluation base…

Language ModelingLanguage ModellingQuestion Answering

How the Softmax Output is Misleading for Evaluating the Strength of Adversarial Examples

2018-11-21 · Utku Ozbulak, Wesley De Neve, Arnout Van Messem

Even before deep learning architectures became the de facto models for complex computer vision tasks, the softmax function was, given its elegant properties, already used to analyze the predictions of feedforward neural …