paper-with-me

Papers

Measuring Context-Word Biases in Lexical Semantic Datasets

2021-12-13 · Qianchu Liu, Diana McCarthy, Anna Korhonen

State-of-the-art pretrained contextualized models (PCM) eg. BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model represents the coupled word and context semantics. We question this assumption by presenting the first quantitative analysis on the context-word interaction being tested in major contextual lexical semantic tasks. To achieve this, we run probing baselines on masked input, and propose measures to calculate and visualize the degree of context or word biases in existing datasets. The analysis was performed on both models and humans. Our findings demonstrate that models are usually not being tested for word-in-context semantics in the same way as humans are in these tasks, which helps us better understand the model-human gap. Specifically, to PCMs, most existing datasets fall into the extreme ends (the retrieval-based tasks exhibit strong target word bias while WiC-style tasks and WSD show strong context bias); In comparison, humans are less biased and achieve much better performance when both word and context are available than with masked input. We recommend our framework for understanding and controlling these biases for model interpretation and future task design.

📄 PDF Abstract BibTeX arXiv:2112.06733

Code (0)

등록된 구현이 없습니다.

Tasks

Retrieval

Methods 이 논문이 사용한 방법론

Multi-Head Attention 설명 없음
Attention 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Adam 설명 없음
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Linear Warmup With Linear Decay Linear Warmup With Linear Decay is a learning rate schedule in which we increase the learning rate linearly for $n$ updates and then linearly decay afterwards.
Residual Connection 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…

Similar Papers 제목 키워드 기반

Measuring Word-Context Biases in Lexical Semantic Datasets

2022-01-16 · ACL ARR January 2022 1 · Anonymous

State-of-the-art contextualized models eg. BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model represents…

Retrieval

Context vs Target Word: Quantifying Biases When Applying Models to Lexical Semantic Datasets

2021-09-17 · ACL ARR September 2021 9 · Anonymous

State-of-the-art contextualized models such as BERT use tasks such as WiC and WSD to evaluate their word-in-context representations. This inherently assumes that performance in these tasks reflect how well a model repres…

Entity Linking

Models can use keywords to answer questions that human cannot

2021-11-16 · ACL ARR November 2021 11 · Anonymous

Recent studies raised that reading comprehension (RC) models learn to exploit biases and annotation artifacts in current Machine Reading Comprehension (MRC) datasets to achieve impressive performance. This hinders the co…

Machine Reading ComprehensionReading Comprehension

DiBiMT: A Novel Benchmark for Measuring Word Sense Disambiguation Biases in Machine Translation

2022-05-01 · ACL 2022 5 · Niccolò Campolungo, Federico Martelli, Francesco Saina, Roberto Navigli

Lexical ambiguity poses one of the greatest challenges in the field of Machine Translation. Over the last few decades, multiple efforts have been undertaken to investigate incorrect translations caused by the polysemous …

Machine TranslationTranslationWord Sense Disambiguation

Measuring Inductive Biases of In-Context Learning with Underspecified Demonstrations

2023-05-22 · Chenglei Si, Dan Friedman, Nitish Joshi, Shi Feng 외

In-context learning (ICL) is an important paradigm for adapting large language models (LLMs) to new tasks, but the generalization behavior of ICL remains poorly understood. We investigate the inductive biases of ICL from…

In-Context LearningInductive Bias