Probing structural constraints of negation in Pretrained Language Models
Université Paris Cité（巴黎城市大学）；Université Sorbonne Nouvelle（新索邦大学）；CNRS（法国国家科学研究中心）；ENS-PSL（巴黎高等师范学院-PSL大学）
David Kletz, Marie Candito, Pascal Amsili
发表信息：
24th Nordic Conference on Computational Linguistics (NoDaLiDa), Northern European Association for Language Technology (NEALT), May 2023, T{ó}rshavn, Faroe Islands. pp.541-554
近期关于预训练语言模型（PLMs）对否定语义影响编码的研究得出了相互矛盾的结论（例如 Kassner 和 Schütze（2020）；Gubelmann 和 Handschuh（2022））。本文聚焦于 PLMs 编码否定及其形式影响的方式，具体通过英语中负极项（NPI）许可现象进行研究。更确切地说，我们使用探针来识别哪些上下文表示能够最好地编码：1）句子中否定的存在，以及 2）相邻被掩码极性项的极性。我们发现，否定辖域内词元的上下文表示确实允许：（i）与辖域外的词元相比，更好地预测 not 的存在；（ii）更好地预测由 not 许可的被掩码极性项的正确极性，尽管差异幅度因 PLM 而异。重要的是，在两种情况下，即使控制了与 not 的距离，这一趋势仍然成立。这倾向于表明这些模型的嵌入确实反映了否定辖域的概念，并且确实编码了否定对 NPI 许可的影响。然而，进一步的对照实验揭示，当使用同一句法子句内词元的上下文表示时，其他词汇项的存在也能被更好地捕捉到，而非子句外的表示，这表明 PLMs 可能只是捕捉到了更一般的句法子句概念。
英文摘要：
Contradictory results about the encoding of the semantic impact of negation in pretrained language models (PLMs). have been drawn recently (e.g. Kassner and Sch{ü}tze (2020); Gubelmann and Handschuh (2022)). In this paper we focus rather on the way PLMs encode negation and its formal impact, through the phenomenon of the Negative Polarity Item (NPI) licensing in English. More precisely, we use probes to identify which contextual representations best encode 1) the presence of negation in a sentence, and 2) the polarity of a neighboring masked polarity item. We find that contextual representations of tokens inside the negation scope do allow for (i) a better prediction of the presence of not compared to those outside the scope and (ii) a better prediction of the right polarity of a masked polarity item licensed by not, although the magnitude of the difference varies from PLM to PLM. Importantly, in both cases the trend holds even when controlling for distance to not. This tends to indicate that the embeddings of these models do reflect the notion of negation scope, and do encode the impact of negation on NPI licensing. Yet, further control experiments reveal that the presence of other lexical items is also better captured when using the contextual representation of a token within the same syntactic clause than outside from it, suggesting that PLMs simply capture the more general notion of syntactic clause.
School of Computing and Information Systems（计算与信息系统学院）；The University of Melbourne（墨尔本大学）
Existing work on probing of pretrained language models (LMs) has predominantly focused on sentence-level syntactic tasks. In this paper, we introduce document-level discourse probing to evaluate the ability of pretrained LMs to capture document-level relations. We experiment with 7 pretrained LMs, 4 languages, and 7 discourse probing tasks, and find BART to be overall the best model at capturing discourse -- but only in its encoder, with BERT performing surprisingly well as the baseline model. Across the different models, there are substantial differences in which layers best capture discourse…
Building on Petroni et al. (2019), we propose two new probing tasks analyzing factual knowledge stored in Pretrained Language Models (PLMs). (1) Negation. We find that PLMs do not distinguish between negated ("Birds cannot [MASK]") and non-negated ("Birds can [MASK]") cloze questions. (2) Mispriming. Inspired by priming methods in human psychology, we add "misprimes" to cloze questions ("Talk? Birds can [MASK]"). We find that PLMs are easily distracted by misprimes. These results suggest that PLMs still have a long way to go to adequately learn human-like factual knowledge.
University of Chinese Academy of Sciences（中国科学院大学）；Institute of Software, Chinese Academy of Sciences（中国科学院软件研究所）；Beijing Academy of Artificial Intelligence（北京智源人工智能研究院）
Prompt-based probing has been widely used in evaluating the abilities of pretrained language models (PLMs). Unfortunately, recent studies have discovered such an evaluation may be inaccurate, inconsistent and unreliable. Furthermore, the lack of understanding its inner workings, combined with its wide applicability, has the potential to lead to unforeseen risks for evaluating and applying PLMs in real-world applications. To discover, understand and quantify the risks, this paper investigates the prompt-based probing from a causal view, highlights three critical biases which could induce biased…
University of Cambridge（剑桥大学）；University of Mannheim（曼海姆大学）
The success of large pretrained language models (LMs) such as BERT and RoBERTa has sparked interest in probing their representations, in order to unveil what types of knowledge they implicitly capture. While prior research focused on morphosyntactic, semantic, and world knowledge, it remains unclear to which extent LMs also derive lexical type-level knowledge from words in context. In this work, we present a systematic empirical analysis across six typologically diverse languages and five different lexical tasks, addressing the following questions: 1) How do different lexical knowledge extract…
LMU Munich（慕尼黑大学）；Munich Center of Machine Learning（慕尼黑机器学习中心）；University of Oxford（牛津大学）
Construction Grammar (CxG) is a paradigm from cognitive linguistics emphasising the connection between syntax and semantics. Rather than rules that operate on lexical items, it posits constructions as the central building blocks of language, i.e., linguistic units of different granularity that combine syntax and semantics. As a first step towards assessing the compatibility of CxG with the syntactic and semantic knowledge demonstrated by state-of-the-art pretrained language models (PLMs), we present an investigation of their capability to classify and understand one of the most commonly studie…
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…
Google Research（谷歌研究院）；Seoul National University（首尔国立大学）
Multimodal learning can benefit from the representation power of pretrained Large Language Models (LLMs). However, state-of-the-art transformer based LLMs often ignore negations in natural language and there is no existing benchmark to quantitatively evaluate whether multimodal transformers inherit this weakness. In this study, we present a new multimodal question answering (QA) benchmark adapted from labeled music videos in AudioSet (Gemmeke et al., 2017) with the goal of systematically evaluating if multimodal transformers can perform complex reasoning to recognize new concepts as negation o…
College of Information and Computer Science, University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校信息与计算机科学学院）；Department of Computer Science, University of Massachusetts Lowell（马萨诸塞大学洛厄尔分校计算机科学系）；Department of Medicine, University of Massachusetts Medical School（马萨诸塞大学医学院医学系）；Center for Healthcare Organization and Implementation Research, Bedford Veterans Affairs Medical Center（贝德福德退伍军人事务医疗中心医疗保健组织与实施研究中心）
Pretrained language models (PLMs) have motivated research on what kinds of knowledge these models learn. Fill-in-the-blanks problem (e.g., cloze tests) is a natural approach for gauging such knowledge. BioLAMA generates prompts for biomedical factual knowledge triples and uses the Top-k accuracy metric to evaluate different PLMs' knowledge. However, existing research has shown that such prompt-based knowledge probing methods can only probe a lower bound of knowledge. Many factors like prompt-based probing biases make the LAMA benchmark unreliable and unstable. This problem is more prominent in…
NLP is currently dominated by general-purpose pretrained language models like RoBERTa, which achieve strong performance on NLU tasks through pretraining on billions of words. But what exact knowledge or skills do Transformer LMs learn from large-scale pretraining that they cannot learn from less data? We adopt four probing methods---classifier probing, information-theoretic probing, unsupervised relative acceptability judgment, and fine-tuning on NLU tasks---and draw learning curves that track the growth of these different measures of linguistic ability with respect to pretraining data volume …
Peking University（北京大学）；Shanghai AI Lab（上海人工智能实验室）；University of California, Santa Barbara（加州大学圣巴巴拉分校）
Previous literature has proved that Pretrained Language Models (PLMs) can store factual knowledge. However, we find that facts stored in the PLMs are not always correct. It motivates us to explore a fundamental question: How do we calibrate factual knowledge in PLMs without re-training from scratch? In this work, we propose a simple and lightweight method CaliNet to achieve this goal. To be specific, we first detect whether PLMs can learn the right facts via a contrastive score between right and fake facts. If not, we then use a lightweight method to add and adapt new parameters to specific fa…
Pretrained language models have shown success in many natural language processing tasks. Many works explore incorporating knowledge into language models. In the biomedical domain, experts have taken decades of effort on building large-scale knowledge bases. For example, the Unified Medical Language System (UMLS) contains millions of entities with their synonyms and defines hundreds of relations among entities. Leveraging this knowledge can benefit a variety of downstream tasks such as named entity recognition and relation extraction. To this end, we propose KeBioLM, a biomedical pretrained lan…
University of Cambridge（剑桥大学）；University of Glasgow（格拉斯哥大学）
Acquiring factual knowledge with Pretrained Language Models (PLMs) has attracted increasing attention, showing promising performance in many knowledge-intensive tasks. Their good performance has led the community to believe that the models do possess a modicum of reasoning competence rather than merely memorising the knowledge. In this paper, we conduct a comprehensive evaluation of the learnable deductive (also known as explicit) reasoning capability of PLMs. Through a series of controlled experiments, we posit two main findings. (i) PLMs inadequately generalise learned logic rules and perfor…
Hebrew University of Jerusalem（耶路撒冷希伯来大学）；University of Washington（华盛顿大学）；Apple, Inc.（苹果公司）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）
The attention mechanism is considered the backbone of the widely-used Transformer architecture. It contextualizes the input by computing input-specific attention matrices. We find that this mechanism, while powerful and elegant, is not as important as typically thought for pretrained language models. We introduce PAPA, a new probing method that replaces the input-dependent attention matrices with constant ones -- the average attention weights over multiple inputs. We use PAPA to analyze several established pretrained Transformers on six downstream tasks. We find that without any input-dependen…
Center for Information and Language Processing, LMU Munich（慕尼黑大学信息与语言处理中心）；Munich Center of Machine Learning（慕尼黑机器学习中心）；Language Technologies Institute, Carnegie Mellon University（卡内基梅隆大学语言技术研究所）
Construction Grammar (CxG) has recently been used as the basis for probing studies that have investigated the performance of large pretrained language models (PLMs) with respect to the structure and meaning of constructions. In this position paper, we make suggestions for the continuation and augmentation of this line of research. We look at probing methodology that was not designed with CxG in mind, as well as probing methodology that was designed for specific constructions. We analyse selected previous work in detail, and provide our view of the most important challenges and research questio…
Language models (LMs) have proven surprisingly successful at capturing factual knowledge by completing cloze-style fill-in-the-blank questions such as "Punta Cana is located in _." However, while knowledge is both written and queried in many languages, studies on LMs' factual representation ability have almost invariably been performed on English. To assess factual knowledge retrieval in LMs in different languages, we create a multilingual benchmark of cloze-style probes for 23 typologically diverse languages. To properly handle language variations, we expand probing methods from single- to mu…
The University of Melbourne（墨尔本大学）；MBZUAI（穆罕默德·本·扎耶德人工智能大学）；RMIT University（皇家墨尔本理工大学）
Negation is poorly captured by current language models, although the extent of this problem is not widely understood. We introduce a natural language inference (NLI) test suite to enable probing the capabilities of NLP methods, with the aim of understanding sub-clausal negation. The test suite contains premise--hypothesis pairs where the premise contains sub-clausal negation and the hypothesis is constructed by making minimal modifications to the premise in order to reflect different possible interpretations. Aside from adopting standard NLI labels, our test suite is systematically constructed…
The logical negation property (LNP), which implies generating different predictions for semantically opposite inputs, is an important property that a trustworthy language model must satisfy. However, much recent evidence shows that large-size pre-trained language models (PLMs) do not satisfy this property. In this paper, we perform experiments using probing tasks to assess PLM's LNP understanding. Unlike previous studies that only examined negation expressions, we expand the boundary of the investigation to lexical semantics. Through experiments, we observe that PLMs violate the LNP frequently…
How does language model pretraining help transfer learning? We consider a simple ablation technique for determining the impact of each pretrained layer on transfer task performance. This method, partial reinitialization, involves replacing different layers of a pretrained model with random weights, then finetuning the entire model on the transfer task and observing the change in performance. This technique reveals that in BERT, layers with high probing performance on downstream GLUE tasks are neither necessary nor sufficient for high accuracy on those tasks. Furthermore, the benefit of using p…
University of North Texas（北得克萨斯大学）；University of Arizona（亚利桑那大学）；Amazon（亚马逊）
Negation poses a challenge in many natural language understanding tasks. Inspired by the fact that understanding a negated statement often requires humans to infer affirmative interpretations, in this paper we show that doing so benefits models for three natural language understanding tasks. We present an automated procedure to collect pairs of sentences with negation and their affirmative interpretations, resulting in over 150,000 pairs. Experimental results show that leveraging these pairs helps (a) T5 generate affirmative interpretations from negations in a previous benchmark, and (b) a RoB…
UNC-Chapel Hill（北卡罗来纳大学教堂山分校）；MILA（蒙特利尔学习算法研究所）；NVIDIA（英伟达）；Mila-Quebec AI Institute（米拉-魁北克人工智能研究所）；Carnegie Mellon University（卡内基梅隆大学）；Salesforce Research（赛富时研究院）
Pretrained language models (PLMs) have been shown to accumulate factual knowledge during pretrainingng (Petroni et al., 2019). Recent works probe PLMs for the extent of this knowledge through prompts either in discrete or continuous forms. However, these methods do not consider symmetry of the task: object prediction and subject prediction. In this work, we propose Symmetrical Prompt Enhancement (SPE), a continuous prompt-based method for factual probing in PLMs that leverages the symmetry of the task by constructing symmetrical prompts for subject and object prediction. Our results on a popul…
Language-Agnostic Bias Detection in Language Models with Bias Probing
Center for Information and Language Processing, LMU Munich（慕尼黑大学信息与语言处理中心）；Munich Center for Machine Learning（慕尼黑机器学习中心）；Data Analytics and Computational Social Science, University of Massachusetts Amherst（马萨诸塞大学阿默斯特分校数据分析与计算社会科学系）；Marmara University（马尔马拉大学）；Language Technology Lab, University of Cambridge（剑桥大学语言技术实验室）；Harvard Kennedy School（哈佛大学肯尼迪学院）
Do PLMs Know and Understand Ontological Knowledge?
ShanghaiTech University（上海科技大学）；Shanghai Engineering Research Center of Intelligent Vision and Imaging（上海智能视觉成像工程研究中心）；DAMO Academy, Alibaba Group（阿里巴巴集团达摩院）
How much knowledge do pretrained language models hold? Recent research observed that pretrained transformers are adept at modeling semantics but it is unclear to what degree they grasp human knowledge, or how to ensure they do so. In this paper we incorporate knowledge-awareness in language model pretraining without changing the transformer architecture, inserting explicit knowledge layers, or adding external storage of semantic information. Rather, we simply signal the existence of entities to the input of the transformer in pretraining, with an entity-extended tokenizer; and at the output, w…
Agents that negotiate with humans find broad applications in pedagogy and conversational AI. Most efforts in human-agent negotiations rely on restrictive menu-driven interfaces for communication. To advance the research in language-based negotiation systems, we explore a novel task of early prediction of buyer-seller negotiation outcomes, by varying the fraction of utterances that the model can access. We explore the feasibility of early prediction by using traditional feature-based methods, as well as by incorporating the non-linguistic task context into a pretrained language model using sent…
Amazon AI（亚马逊人工智能）；Cairo University（开罗大学）；Department of Computer Science, Columbia University（哥伦比亚大学计算机科学系）
Dialogue summarization comes with its own peculiar challenges as opposed to news or scientific articles summarization. In this work, we explore four different challenges of the task: handling and differentiating parts of the dialogue belonging to multiple speakers, negation understanding, reasoning about the situation, and informal language understanding. Using a pretrained sequence-to-sequence language model, we explore speaker name substitution, negation scope highlighting, multi-task learning with relevant tasks, and pretraining on in-domain data. Our experiments show that our proposed tech…
Pretrained language models (PLMs) often fail to fairly represent target users from certain world regions because of the under-representation of those regions in training datasets. With recent PLMs trained on enormous data sources, quantifying their potential biases is difficult, due to their black-box nature and the sheer scale of the data sources. In this work, we devise an approach to study the geographic bias (and knowledge) present in PLMs, proposing a Geographic-Representation Probing Framework adopting a self-conditioning method coupled with entity-country mappings. Our findings suggest …
Siebel School of Computing and Data Science（西贝尔计算与数据科学学院）；The Grainger College of Engineering（格兰杰工程学院）；University of Illinois Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
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…
Term clustering is important in biomedical knowledge graph construction. Using similarities between terms embedding is helpful for term clustering. State-of-the-art term embeddings leverage pretrained language models to encode terms, and use synonyms and relation knowledge from knowledge graphs to guide contrastive learning. These embeddings provide close embeddings for terms belonging to the same concept. However, from our probing experiments, these embeddings are not sensitive to minor textual differences which leads to failure for biomedical term clustering. To alleviate this problem, we ad…
How does the input segmentation of pretrained language models (PLMs) affect their interpretations of complex words? We present the first study investigating this question, taking BERT as the example PLM and focusing on its semantic representations of English derivatives. We show that PLMs can be interpreted as serial dual-route models, i.e., the meanings of complex words are either stored or else need to be computed from the subwords, which implies that maximally meaningful input tokens should allow for the best generalization on new words. This hypothesis is confirmed by a series of semantic …
Deep learning models are widely used for solving challenging code processing tasks, such as code generation or code summarization. Traditionally, a specific model architecture was carefully built to solve a particular code processing task. However, recently general pretrained models such as CodeBERT or CodeT5 have been shown to outperform task-specific models in many applications. While pretrained models are known to learn complex patterns from data, they may fail to understand some properties of source code. To test diverse aspects of code understanding, we introduce a set of diagnosting prob…
New York University（纽约大学）；University of Colorado Boulder（科罗拉多大学博尔德分校）
While pretrained models such as BERT have shown large gains across natural language understanding tasks, their performance can be improved by further training the model on a data-rich intermediate task, before fine-tuning it on a target task. However, it is still poorly understood when and why intermediate-task training is beneficial for a given target task. To investigate this, we perform a large-scale study on the pretrained RoBERTa model with 110 intermediate-target task combinations. We further evaluate all trained models with 25 probing tasks meant to reveal the specific skills that drive…
Carnegie Mellon University（卡内基梅隆大学）；UC San Diego（加州大学圣迭戈分校）
Physical measurements constitute a large portion of numbers in academic papers, engineering reports, and web tables. Current benchmarks fall short of properly evaluating numeracy of pretrained language models on measurements, hindering research on developing new methods and applying them to numerical tasks. To that end, we introduce a novel task, Masked Measurement Prediction (MMP), where a model learns to reconstruct a number together with its associated unit given masked text. MMP is useful for both training new numerically informed models as well as evaluating numeracy of existing systems. …
New York University（纽约大学）；Google AI Language（谷歌人工智能语言团队）；Harvard University（哈佛大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；Brown University（布朗大学）
We introduce a set of nine challenge tasks that test for the understanding of function words. These tasks are created by structurally mutating sentences from existing datasets to target the comprehension of specific types of function words (e.g., prepositions, wh-words). Using these probing tasks, we explore the effects of various pretraining objectives for sentence encoders (e.g., language modeling, CCG supertagging and natural language inference (NLI)) on the learned representations. Our results show that pretraining on language modeling performs the best on average across our probing tasks,…
University of Edinburgh（爱丁堡大学）；University of Amsterdam（阿姆斯特丹大学）
To measure how well pretrained representations encode some linguistic property, it is common to use accuracy of a probe, i.e. a classifier trained to predict the property from the representations. Despite widespread adoption of probes, differences in their accuracy fail to adequately reflect differences in representations. For example, they do not substantially favour pretrained representations over randomly initialized ones. Analogously, their accuracy can be similar when probing for genuine linguistic labels and probing for random synthetic tasks. To see reasonable differences in accuracy wi…
International Institute of Information Technology, Bangalore（班加罗尔国际信息技术学院）；Indraprastha Institute of Information Technology, Delhi（德里印德拉普拉斯塔信息技术学院）；International Institute of Information Technology, Hyderabad（海得拉巴国际信息技术学院）
Online news and social media have been the de facto mediums to disseminate information globally from the beginning of the last decade. However, bias in content and purpose of intentions are not regulated, and managing bias is the responsibility of content consumers. In this regard, understanding the stances and biases of news sources towards specific entities becomes important. To address this problem, we use pretrained language models, which have been shown to bring about good results with no task-specific training or few-shot training. In this work, we approach the problem of characterizing …
John A. Paulson School of Engineering and Applied Sciences（约翰·A·保尔森工程与应用科学学院）；Harvard University（哈佛大学）
Pretrained large language models (LLMs) are becoming increasingly powerful and ubiquitous in mainstream applications such as being a personal assistant, a dialogue model, etc. As these models become proficient in deducing user preferences and offering tailored assistance, there is an increasing concern about the ability of these models to influence, modify and in the extreme case manipulate user preference adversarially. The issue of lack of interpretability in these models in adversarial settings remains largely unsolved. This work tries to study adversarial behavior in user preferences from …
