Although models using contextual word embeddings have achieved state-of-the-art results on a host of NLP tasks, little is known about exactly what information these embeddings encode about the context words that they are understood to reflect. To address this question, we introduce a suite of probing tasks that enable fine-grained testing of contextual embeddings for encoding of information about surrounding words. We apply these tasks to examine the popular BERT, ELMo and GPT contextual encoders, and find that each of our tested information types is indeed encoded as contextual information across tokens, often with near-perfect recoverability-but the encoders vary in which features they distribute to which tokens, how nuanced their distributions are, and how robust the encoding of each feature is to distance. We discuss implications of these results for how different types of models breakdown and prioritize word-level context information when constructing token embeddings.
Google AI Language；Johns Hopkins University（约翰斯·霍普金斯大学）；Swarthmore College（斯沃斯莫尔学院）；New York University（纽约大学）；Brown University（布朗大学）
Contextualized representation models such as ELMo (Peters et al., 2018a) and BERT (Devlin et al., 2018) have recently achieved state-of-the-art results on a diverse array of downstream NLP tasks. Building on recent token-level probing work, we introduce a novel edge probing task design and construct a broad suite of sub-sentence tasks derived from the traditional structured NLP pipeline. We probe word-level contextual representations from four recent models and investigate how they encode sentence structure across a range of syntactic, semantic, local, and long-range phenomena. We find that ex…
Pretrained language models have achieved a new state of the art on many NLP tasks, but there are still many open questions about how and why they work so well. We investigate the contextualization of words in BERT. We quantify the amount of contextualization, i.e., how well words are interpreted in context, by studying the extent to which semantic classes of a word can be inferred from its contextualized embeddings. Quantifying contextualization helps in understanding and utilizing pretrained language models. We show that top layer representations achieve high accuracy inferring semantic class…
Boston University（波士顿大学）；University of Pittsburgh（匹兹堡大学）；ETH Zurich（苏黎世联邦理工学院）
We propose reconstruction probing, a new analysis method for contextualized representations based on reconstruction probabilities in masked language models (MLMs). This method relies on comparing the reconstruction probabilities of tokens in a given sequence when conditioned on the representation of a single token that has been fully contextualized and when conditioned on only the decontextualized lexical prior of the model. This comparison can be understood as quantifying the contribution of contextualization towards reconstruction -- the difference in the reconstruction probabilities can onl…
As the name implies, contextualized representations of language are typically motivated by their ability to encode context. Which aspects of context are captured by such representations? We introduce an approach to address this question using Representational Similarity Analysis (RSA). As case studies, we investigate the degree to which a verb embedding encodes the verb's subject, a pronoun embedding encodes the pronoun's antecedent, and a full-sentence representation encodes the sentence's head word (as determined by a dependency parse). In all cases, we show that BERT's contextualized embedd…
Paul G. Allen School of Computer Science & Engineering（保罗·G·艾伦计算机科学与工程学院）；University of Washington（华盛顿大学）
The success of large-scale contextual language models has attracted great interest in probing what is encoded in their representations. In this work, we consider a new question: to what extent contextual representations of concrete nouns are aligned with corresponding visual representations? We design a probing model that evaluates how effective are text-only representations in distinguishing between matching and non-matching visual representations. Our findings show that language representations alone provide a strong signal for retrieving image patches from the correct object categories. Mor…
Alibaba Group（阿里巴巴集团）；Beijing Jiaotong University（北京交通大学）；University of Edinburgh（爱丁堡大学）
Recently, a variety of probing tasks are proposed to discover linguistic properties learned in contextualized word embeddings. Many of these works implicitly assume these embeddings lay in certain metric spaces, typically the Euclidean space. This work considers a family of geometrically special spaces, the hyperbolic spaces, that exhibit better inductive biases for hierarchical structures and may better reveal linguistic hierarchies encoded in contextualized representations. We introduce a Poincare probe, a structural probe projecting these embeddings into a Poincare subspace with explicitly …
Probing experiments investigate the extent to which neural representations make properties -- like part-of-speech -- predictable. One suggests that a representation encodes a property if probing that representation produces higher accuracy than probing a baseline representation like non-contextual word embeddings. Instead of using baselines as a point of comparison, we're interested in measuring information that is contained in the representation but not in the baseline. For example, current methods can detect when a representation is more useful than the word identity (a baseline) for predict…
We investigate the extent to which verb alternation classes, as described by Levin (1993), are encoded in the embeddings of Large Pre-trained Language Models (PLMs) such as BERT, RoBERTa, ELECTRA, and DeBERTa using selectively constructed diagnostic classifiers for word and sentence-level prediction tasks. We follow and expand upon the experiments of Kann et al. (2019), which aim to probe whether static embeddings encode frame-selectional properties of verbs. At both the word and sentence level, we find that contextual embeddings from PLMs not only outperform non-contextual embeddings, but ach…
University of Pittsburgh（匹兹堡大学）；Carnegie Mellon University（卡内基梅隆大学）；Google, Inc.（谷歌公司）
Contextualized word embeddings derived from pre-trained language models (LMs) show significant improvements on downstream NLP tasks. Pre-training on domain-specific corpora, such as biomedical articles, further improves their performance. In this paper, we conduct probing experiments to determine what additional information is carried intrinsically by the in-domain trained contextualized embeddings. For this we use the pre-trained LMs as fixed feature extractors and restrict the downstream task models to not have additional sequence modeling layers. We compare BERT, ELMo, BioBERT and BioELMo, …
Pre-trained contextual representations have led to dramatic performance improvements on a range of downstream tasks. Such performance improvements have motivated researchers to quantify and understand the linguistic information encoded in these representations. In general, researchers quantify the amount of linguistic information through probing, an endeavor which consists of training a supervised model to predict a linguistic property directly from the contextual representations. Unfortunately, this definition of probing has been subject to extensive criticism in the literature, and has been …
Sentinel: Decoding Context Utilization via Attention Probing for Efficient LLM Context Compression
Ping An Technology (Shenzhen) Co., Ltd., China（平安科技（深圳）有限公司，中国）；University of Science and Technology of China（中国科学技术大学）；University of Electronic Science and Technology of China（电子科技大学）
ADAPT Centre（ADAPT中心）；Technological University Dublin（都柏林理工大学）
Improving our understanding of how information is encoded in vector space can yield valuable interpretability insights. Alongside vector dimensions, we argue that it is possible for the vector norm to also carry linguistic information. We develop a method to test this: an extension of the probing framework which allows for relative intrinsic interpretations of probing results. It relies on introducing noise that ablates information encoded in embeddings, grounded in random baselines and confidence intervals. We apply the method to well-established probing tasks and find evidence that confirms …
NLP has a rich history of representing our prior understanding of language in the form of graphs. Recent work on analyzing contextualized text representations has focused on hand-designed probe models to understand how and to what extent do these representations encode a particular linguistic phenomenon. However, due to the inter-dependence of various phenomena and randomness of training probe models, detecting how these representations encode the rich information in these linguistic graphs remains a challenging problem. In this paper, we propose a new information-theoretic probe, Bird's Eye, …
Beyond Black-Box Interventions: Latent Probing for Faithful Retrieval-Augmented Generation
School of Informatics, Xiamen University（厦门大学信息学院）；The Hong Kong Polytechnic University（香港理工大学）；University of Chinese Academy of Sciences（中国科学院大学）；Alibaba Group（阿里巴巴集团）；Key Laboratory of Digital Protection and Intelligent Processing of Intangible Cultural Heritage of Fujian and Taiwan (Xiamen University), Ministry of Culture and Tourism, China（福建省和台湾非物质文化遗产数字化保护与智能处理重点实验室（厦门大学），中华人民共和国文化和旅游部，中国）
Progress in pre-trained language models has led to a surge of impressive results on downstream tasks for natural language understanding. Recent work on probing pre-trained language models uncovered a wide range of linguistic properties encoded in their contextualized representations. However, it is unclear whether they encode semantic knowledge that is crucial to symbolic inference methods. We propose a methodology for probing linguistic information for logical inference in pre-trained language model representations. Our probing datasets cover a list of linguistic phenomena required by major s…
Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；University of California, Los Angeles（加州大学洛杉矶分校）；University of Virginia（弗吉尼亚大学）；University of Cambridge（剑桥大学）
In this paper, we quantify, analyze and mitigate gender bias exhibited in ELMo's contextualized word vectors. First, we conduct several intrinsic analyses and find that (1) training data for ELMo contains significantly more male than female entities, (2) the trained ELMo embeddings systematically encode gender information and (3) ELMo unequally encodes gender information about male and female entities. Then, we show that a state-of-the-art coreference system that depends on ELMo inherits its bias and demonstrates significant bias on the WinoBias probing corpus. Finally, we explore two methods …
Québec Artificial Intelligence Institute (Mila)（魁北克人工智能研究所（米拉））；University of Cambridge（剑桥大学）；Facebook AI Research（脸书人工智能研究院）；ETH Zürich（苏黎世联邦理工学院）
Most modern NLP systems make use of pre-trained contextual representations that attain astonishingly high performance on a variety of tasks. Such high performance should not be possible unless some form of linguistic structure inheres in these representations, and a wealth of research has sprung up on probing for it. In this paper, we draw a distinction between intrinsic probing, which examines how linguistic information is structured within a representation, and the extrinsic probing popular in prior work, which only argues for the presence of such information by showing that it can be succes…
Pretraining deep language models has led to large performance gains in NLP. Despite this success, Schick and Schütze (2020) recently showed that these models struggle to understand rare words. For static word embeddings, this problem has been addressed by separately learning representations for rare words. In this work, we transfer this idea to pretrained language models: We introduce BERTRAM, a powerful architecture based on BERT that is capable of inferring high-quality embeddings for rare words that are suitable as input representations for deep language models. This is achieved by enabling…
Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；University of California, Irvine（加州大学欧文分校）；University of Washington（华盛顿大学）
Contextual word representations, typically trained on unstructured, unlabeled text, do not contain any explicit grounding to real world entities and are often unable to remember facts about those entities. We propose a general method to embed multiple knowledge bases (KBs) into large scale models, and thereby enhance their representations with structured, human-curated knowledge. For each KB, we first use an integrated entity linker to retrieve relevant entity embeddings, then update contextual word representations via a form of word-to-entity attention. In contrast to previous approaches, the…
Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；Harvard John A. Paulson School of Engineering and Applied Sciences（哈佛大学约翰·A·保尔森工程与应用科学学院）；MIT Computer Science and Artificial Intelligence Laboratory（麻省理工学院计算机科学与人工智能实验室）；University of Washington（华盛顿大学）
Contextual word representations derived from large-scale neural language models are successful across a diverse set of NLP tasks, suggesting that they encode useful and transferable features of language. To shed light on the linguistic knowledge they capture, we study the representations produced by several recent pretrained contextualizers (variants of ELMo, the OpenAI transformer language model, and BERT) with a suite of seventeen diverse probing tasks. We find that linear models trained on top of frozen contextual representations are competitive with state-of-the-art task-specific models in…
Fine-tuning pre-trained contextualized embedding models has become an integral part of the NLP pipeline. At the same time, probing has emerged as a way to investigate the linguistic knowledge captured by pre-trained models. Very little is, however, understood about how fine-tuning affects the representations of pre-trained models and thereby the linguistic knowledge they encode. This paper contributes towards closing this gap. We study three different pre-trained models: BERT, RoBERTa, and ALBERT, and investigate through sentence-level probing how fine-tuning affects their representations. We …
Understanding how linguistic structures are encoded in contextualized embedding could help explain their impressive performance across NLP@. Existing approaches for probing them usually call for training classifiers and use the accuracy, mutual information, or complexity as a proxy for the representation's goodness. In this work, we argue that doing so can be unreliable because different representations may need different classifiers. We develop a heuristic, DirectProbe, that directly studies the geometry of a representation by building upon the notion of a version space for a task. Experiment…
Contextual representations learned by language models can often encode undesirable attributes, like demographic associations of the users, while being trained for an unrelated target task. We aim to scrub such undesirable attributes and learn fair representations while maintaining performance on the target task. In this paper, we present an adversarial learning framework "Adversarial Scrubber" (ADS), to debias contextual representations. We perform theoretical analysis to show that our framework converges without leaking demographic information under certain conditions. We extend previous eval…
Intercepting the Kangaroo: Experimental Astrolinguistics with Constructed Lexicons, Active Probing, and Large Language Models as Informants and Hypothesis Proposers
INESC TEC（葡萄牙系统与计算机工程、技术与科学研究所）；University of Porto（波尔图大学）；Cardiff University（卡迪夫大学）
Distributional semantics based on neural approaches is a cornerstone of Natural Language Processing, with surprising connections to human meaning representation as well. Recent Transformer-based Language Models have proven capable of producing contextual word representations that reliably convey sense-specific information, simply as a product of self-supervision. Prior work has shown that these contextual representations can be used to accurately represent large sense inventories as sense embeddings, to the extent that a distance-based solution to Word Sense Disambiguation (WSD) tasks outperfo…
University of Cambridge（剑桥大学）；ETH Zürich（苏黎世联邦理工学院）
In this paper, we seek to measure how much information a component in a neural network could extract from the representations fed into it. Our work stands in contrast to prior probing work, most of which investigates how much information a model's representations contain. This shift in perspective leads us to propose a new principle for probing, the architectural bottleneck principle: In order to estimate how much information a given component could extract, a probe should look exactly like the component. Relying on this principle, we estimate how much syntactic information is available to tra…
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 dat…
Recent works show that learning contextualized embeddings for words is beneficial for downstream tasks. BERT is one successful example of this approach. It learns embeddings by solving two tasks, which are masked language model (masked LM) and the next sentence prediction (NSP). The pre-training of BERT can also be framed as a multitask learning problem. In this work, we adopt hierarchical multitask learning approaches for BERT pre-training. Pre-training tasks are solved at different layers instead of the last layer, and information from the NSP task is transferred to the masked LM task. Also,…
We present a novel deep learning-based framework to generate embedding representations of fine-grained emotions that can be used to computationally describe psychological models of emotions. Our framework integrates a contextualized embedding encoder with a multi-head probing model that enables to interpret dynamically learned representations optimized for an emotion classification task. Our model is evaluated on the Empathetic Dialogue dataset and shows the state-of-the-art result for classifying 32 emotions. Our layer analysis can derive an emotion graph to depict hierarchical relations amon…
Sentence embedding is an important research topic in natural language processing (NLP) since it can transfer knowledge to downstream tasks. Meanwhile, a contextualized word representation, called BERT, achieves the state-of-the-art performance in quite a few NLP tasks. Yet, it is an open problem to generate a high quality sentence representation from BERT-based word models. It was shown in previous study that different layers of BERT capture different linguistic properties. This allows us to fusion information across layers to find better sentence representation. In this work, we study the lay…
Given the prevalence of pre-trained contextualized representations in today's NLP, there have been many efforts to understand what information they contain, and why they seem to be universally successful. The most common approach to use these representations involves fine-tuning them for an end task. Yet, how fine-tuning changes the underlying embedding space is less studied. In this work, we study the English BERT family and use two probing techniques to analyze how fine-tuning changes the space. We hypothesize that fine-tuning affects classification performance by increasing the distances be…
