Pre-trained language models (PLMs) like BERT are being used for almost all language-related tasks, but interpreting their behavior still remains a significant challenge and many important questions remain largely unanswered. In this work, we re-purpose a sentence editing dataset, where faithful high-quality human rationales can be automatically extracted and compared with extracted model rationales, as a new testbed for interpretability. This enables us to conduct a systematic investigation on an array of questions regarding PLMs' interpretability, including the role of pre-training procedure, comparison of rationale extraction methods, and different layers in the PLM. The investigation generates new insights, for example, contrary to the common understanding, we find that attention weights correlate well with human rationales and work better than gradient-based saliency in extracting model rationales. Both the dataset and code are available at https://github.com/samuelstevens/sentence-editing-interpretability to facilitate future interpretability research.
Seoul National University（首尔国立大学）；AIRS Company, Hyundai Motor Group（现代汽车集团AIRS公司）；Adobe Research（奥多比研究院）
Despite the recent advancements in abstractive summarization systems leveraged from large-scale datasets and pre-trained language models, the factual correctness of the summary is still insufficient. One line of trials to mitigate this problem is to include a post-editing process that can detect and correct factual errors in the summary. In building such a post-editing system, it is strongly required that 1) the process has a high success rate and interpretability and 2) has a fast running time. Previous approaches focus on regeneration of the summary using the autoregressive models, which lac…
Potential and Challenges of Model Editing for Social Debiasing
Zhejiang University School of Engineering（浙江大学工程学院）；Westlake University（西湖大学）；Institute of Advanced Technology, Westlake Institute for Advanced Study（西湖高等研究院先进技术研究所）
University of California, Davis（加州大学戴维斯分校）；ByteDance（字节跳动）
Many social media news writers are not professionally trained. Therefore, social media platforms have to hire professional editors to adjust amateur headlines to attract more readers. We propose to automate this headline editing process through neural network models to provide more immediate writing support for these social media news writers. To train such a neural headline editing model, we collected a dataset which contains articles with original headlines and professionally edited headlines. However, it is expensive to collect a large number of professionally edited headlines. To solve thi…
Though offering amazing contextualized token-level representations, current pre-trained language models take less attention on accurately acquiring sentence-level representation during their self-supervised pre-training. However, contrastive objectives which dominate the current sentence representation learning bring little linguistic interpretability and no performance guarantee on downstream semantic tasks. We instead propose a novel generative self-supervised learning objective based on phrase reconstruction. To overcome the drawbacks of previous generative methods, we carefully model intra…
Finding and Editing Multi-Modal Neurons in Pre-Trained Transformers
University of Science and Technology of China（中国科学技术大学）；School of Computer Science, Fudan University（复旦大学计算机学院）；Tsinghua University（清华大学）；Hefei University of Technology（合肥工业大学）
School of Computer Science and Technology, Beijing Institute of Technology, Beijing, China（北京理工大学计算机科学与技术学院）；School of Computer Science and Engineering, Beihang University, Beijing, China（北京航空航天大学计算机科学与工程学院）
Microsoft Research（微软研究院）；Microsoft Research Asia（微软亚洲研究院）
In this paper, we introduce a novel natural language generation task, termed as text morphing, which targets at generating the intermediate sentences that are fluency and smooth with the two input sentences. We propose the Morphing Networks consisting of the editing vector generation networks and the sentence editing networks which are trained jointly. Specifically, the editing vectors are generated with a recurrent neural networks model from the lexical gap between the source sentence and the target sentence. Then the sentence editing networks iteratively generate new sentences with the curre…
We propose a framework for computer-assisted text editing. It applies to translation post-editing and to paraphrasing. Our proposal relies on very simple interactions: a human editor modifies a sentence by marking tokens they would like the system to change. Our model then generates a new sentence which reformulates the initial sentence by avoiding marked words. The approach builds upon neural sequence-to-sequence modeling and introduces a neural network which takes as input a sentence along with change markers. Our model is trained on translation bitext by simulating post-edits. We demonstrat…
Microsoft Research（微软研究院）；Beihang University（北京航空航天大学）
An intuitive way for a human to write paraphrase sentences is to replace words or phrases in the original sentence with their corresponding synonyms and make necessary changes to ensure the new sentences are fluent and grammatically correct. We propose a novel approach to modeling the process with dictionary-guided editing networks which effectively conduct rewriting on the source sentence to generate paraphrase sentences. It jointly learns the selection of the appropriate word level and phrase level paraphrase pairs in the context of the original sentence from an off-the-shelf dictionary as w…
Better Call SAUL: Fluent and Consistent Language Model Editing with Generation Regularization
Bosch Center for Artificial Intelligence (BCAI)（博世人工智能中心）；LMU Munich（慕尼黑大学）；Munich Center for Machine Learning (MCML)（慕尼黑机器学习中心）；Hochschule der Medien, Stuttgart（斯图加特媒体大学）；Karlsruhe University of Applied Sciences（卡尔斯鲁厄应用技术大学）
We present the first sentence simplification model that learns explicit edit operations (ADD, DELETE, and KEEP) via a neural programmer-interpreter approach. Most current neural sentence simplification systems are variants of sequence-to-sequence models adopted from machine translation. These methods learn to simplify sentences as a byproduct of the fact that they are trained on complex-simple sentence pairs. By contrast, our neural programmer-interpreter is directly trained to predict explicit edit operations on targeted parts of the input sentence, resembling the way that humans might perfor…
The Australian National University（澳大利亚国立大学）；DATA61-CSIRO（澳大利亚联邦科学与工业研究组织 Data61）；Huawei Cyberverse Lab（华为河图实验室）；SenseTime（商汤科技）
This work studies the task of glossification, of which the aim is to em transcribe natural spoken language sentences for the Deaf (hard-of-hearing) community to ordered sign language glosses. Previous sequence-to-sequence language models trained with paired sentence-gloss data often fail to capture the rich connections between the two distinct languages, leading to unsatisfactory transcriptions. We observe that despite different grammars, glosses effectively simplify sentences for the ease of deaf communication, while sharing a large portion of vocabulary with sentences. This has motivated us …
ACE: Attribution-Controlled Knowledge Editing for Multi-hop Factual Recall
HKUST(GZ) Deep Interdisciplinary Intelligence Lab（香港科技大学(广州)深 interdisciplinary 智能实验室）；BUAA（北京航空航天大学）；HKUST(GZ) Institute of Deep Perception Technology, JITRI Deep Interdisciplinary Intelligence Lab（香港科技大学(广州)深度感知技术研究所，JITRI 深 interdisciplinary 智能实验室）
Data augmentation promises to alleviate data scarcity. This is most important in cases where the initial data is in short supply. This is, for existing methods, also where augmenting is the most difficult, as learning the full data distribution is impossible. For natural language, sentence editing offers a solution - relying on small but meaningful changes to the original ones. Learning which changes are meaningful also requires large amounts of training data. We thus aim to learn this in a source domain where data is abundant and apply it in a different, target domain, where data is scarce - …
Instituto de Telecomunicações（电信研究所）；Facebook AI（脸书人工智能研究院）；IQT Labs（IQT实验室）；University of Sheffield（谢菲尔德大学）；Imperial College London（伦敦帝国学院）；Unbabel（Unbabel公司）；Johns Hopkins University（约翰霍普金斯大学）；University of Wolverhampton（伍尔弗汉普顿大学）
We present MLQE-PE, a new dataset for Machine Translation (MT) Quality Estimation (QE) and Automatic Post-Editing (APE). The dataset contains eleven language pairs, with human labels for up to 10,000 translations per language pair in the following formats: sentence-level direct assessments and post-editing effort, and word-level good/bad labels. It also contains the post-edited sentences, as well as titles of the articles where the sentences were extracted from, and the neural MT models used to translate the text.
We present IntelliCAT, an interactive translation interface with neural models that streamline the post-editing process on machine translation output. We leverage two quality estimation (QE) models at different granularities: sentence-level QE, to predict the quality of each machine-translated sentence, and word-level QE, to locate the parts of the machine-translated sentence that need correction. Additionally, we introduce a novel translation suggestion model conditioned on both the left and right contexts, providing alternatives for specific words or phrases for correction. Finally, with wor…
Locate-then-edit for Multi-hop Factual Recall under Knowledge Editing
Peking University（北京大学）；Provable Responsible AI and Data Analytics (PRADA) Lab（可验证负责任人工智能与数据分析（PRADA）实验室）；King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；South China University of Technology（华南理工大学）
Beyond the Sentence: A Survey on Context-Aware Machine Translation with Large Language Models
Department of Computer Science and Engineering, Indian Institute of Technology Patna, India（印度理工学院帕纳瓦分校计算机科学与工程系）；Wipro AI, Lab45, London, UK（Wipro AI，Lab45，伦敦，英国）；School of AI and Data Science, Indian Institute of Technology Jodhpur, India（印度理工学院朱罗普分校人工智能与数据科学学院）
Don't Believe Everything You Read: Enhancing Summarization Interpretability through Automatic Identification of Hallucinations in Large Language Models
We propose a new generative model of sentences that first samples a prototype sentence from the training corpus and then edits it into a new sentence. Compared to traditional models that generate from scratch either left-to-right or by first sampling a latent sentence vector, our prototype-then-edit model improves perplexity on language modeling and generates higher quality outputs according to human evaluation. Furthermore, the model gives rise to a latent edit vector that captures interpretable semantics such as sentence similarity and sentence-level analogies.
While neural sequence generation models achieve initial success for many NLP applications, the canonical decoding procedure with left-to-right generation order (i.e., autoregressive) in one-pass can not reflect the true nature of human revising a sentence to obtain a refined result. In this work, we propose XL-Editor, a novel training framework that enables state-of-the-art generalized autoregressive pretraining methods, XLNet specifically, to revise a given sentence by the variable-length insertion probability. Concretely, XL-Editor can (1) estimate the probability of inserting a variable-len…
University of Technology Sydney（悉尼科技大学）；Capital Markets Cooperative Research Centre（资本市场合作研究中心）
Automatic post-editing (APE) systems aim to correct the systematic errors made by machine translators. In this paper, we propose a neural APE system that encodes the source (src) and machine translated (mt) sentences with two separate encoders, but leverages a shared attention mechanism to better understand how the two inputs contribute to the generation of the post-edited (pe) sentences. Our empirical observations have showed that when the mt is incorrect, the attention shifts weight toward tokens in the src sentence to properly edit the incorrect translation. The model has been trained and e…
Sentence level quality estimation (QE) for machine translation (MT) attempts to predict the translation edit rate (TER) cost of post-editing work required to correct MT output. We describe our view on sentence-level QE as dictated by several practical setups encountered in the industry. We find consumers of MT output---whether human or algorithmic ones---to be primarily interested in a binary quality metric: is the translated sentence adequate as-is or does it need post-editing? Motivated by this we propose a quality classification (QC) view on sentence-level QE whereby we focus on maximizing …
Charles University（查理大学）；Faculty of Mathematics and Physics（数学与物理学院）；Institute of Formal and Applied Linguistics（形式与应用语言学研究所）
We present a novel approach to data-to-text generation based on iterative text editing. Our approach maximizes the completeness and semantic accuracy of the output text while leveraging the abilities of recent pre-trained models for text editing (LaserTagger) and language modeling (GPT-2) to improve the text fluency. To this end, we first transform data items to text using trivial templates, and then we iteratively improve the resulting text by a neural model trained for the sentence fusion task. The output of the model is filtered by a simple heuristic and reranked with an off-the-shelf pre-t…
Previous research on word embeddings has shown that sparse representations, which can be either learned on top of existing dense embeddings or obtained through model constraints during training time, have the benefit of increased interpretability properties: to some degree, each dimension can be understood by a human and associated with a recognizable feature in the data. In this paper, we transfer this idea to sentence embeddings and explore several approaches to obtain a sparse representation. We further introduce a novel, quantitative and automated evaluation metric for sentence embedding i…
Tohoku University（东北大学）；Langsmith Inc.（Langsmith公司）；Yahoo Japan Corporation（雅虎日本公司）；RIKEN（理化学研究所）；Octanove Labs LLC（Octanove实验室有限责任公司）
The writing process consists of several stages such as drafting, revising, editing, and proofreading. Studies on writing assistance, such as grammatical error correction (GEC), have mainly focused on sentence editing and proofreading, where surface-level issues such as typographical, spelling, or grammatical errors should be corrected. We broaden this focus to include the earlier revising stage, where sentences require adjustment to the information included or major rewriting and propose Sentence-level Revision (SentRev) as a new writing assistance task. Well-performing systems in this task ca…
We propose LaserTagger - a sequence tagging approach that casts text generation as a text editing task. Target texts are reconstructed from the inputs using three main edit operations: keeping a token, deleting it, and adding a phrase before the token. To predict the edit operations, we propose a novel model, which combines a BERT encoder with an autoregressive Transformer decoder. This approach is evaluated on English text on four tasks: sentence fusion, sentence splitting, abstractive summarization, and grammar correction. LaserTagger achieves new state-of-the-art results on three of these t…
An interpretability framework for Similar case matching
Guangdong University of Technology（广东工业大学）；KTH Royal Institute of Technology（KTH皇家理工学院）；Guangdong University of Foreign Studies（广东外语外贸大学）；Lingnan Normal University（岭南师范学院）
Tencent AI Lab（腾讯人工智能实验室）；The Chinese University of Hong Kong（香港中文大学）；Noah’s Ark Lab, Huawei Technologies（华为诺亚方舟实验室）
We propose the task of Quantifiable Sequence Editing (QuaSE): editing an input sequence to generate an output sequence that satisfies a given numerical outcome value measuring a certain property of the sequence, with the requirement of keeping the main content of the input sequence. For example, an input sequence could be a word sequence, such as review sentence and advertisement text. For a review sentence, the outcome could be the review rating; for an advertisement, the outcome could be the click-through rate. The major challenge in performing QuaSE is how to perceive the outcome-related wo…
University of California, Santa Cruz（加州大学圣克鲁兹分校）；Lawrence Berkeley National Laboratory（劳伦斯伯克利国家实验室）；University of California, Berkeley（加州大学伯克利分校）
Research Replication Prediction (RRP) is the task of predicting whether a published research result can be replicated or not. Building an interpretable neural text classifier for RRP promotes the understanding of why a research paper is predicted as replicable or non-replicable and therefore makes its real-world application more reliable and trustworthy. However, the prior works on model interpretation mainly focused on improving the model interpretability at the word/phrase level, which are insufficient especially for long research papers in RRP. Furthermore, the existing methods cannot utili…
Pronouns are important determinants of a text's meaning but difficult to translate. This is because pronoun choice can depend on entities described in previous sentences, and in some languages pronouns may be dropped when the referent is inferrable from the context. These issues can lead Neural Machine Translation (NMT) systems to make critical errors on pronouns that impair intelligibility and even reinforce gender bias. We investigate the severity of this pronoun issue, showing that (1) in some domains, pronoun choice can account for more than half of a NMT systems' errors, and (2) pronouns …
SenseTime（商汤科技）；National University of Defense Technology（国防科技大学）；Singapore Management University（新加坡管理大学）
Document-level natural language inference (DOCNLI) is a new challenging task in natural language processing, aiming at judging the entailment relationship between a pair of hypothesis and premise documents. Current datasets and baselines largely follow sentence-level settings, but fail to address the issues raised by longer documents. In this paper, we establish a general solution, named Retrieval, Reading and Fusion (R2F) framework, and a new setting, by analyzing the main challenges of DOCNLI: interpretability, long-range dependency, and cross-sentence inference. The basic idea of the framew…
Data building for automatic post-editing (APE) requires extensive and expert-level human effort, as it contains an elaborate process that involves identifying errors in sentences and providing suitable revisions. Hence, we develop a self-supervised data generation tool, deployable as a web application, that minimizes human supervision and constructs personalized APE data from a parallel corpus for several language pairs with English as the target language. Data-centric APE research can be conducted using this tool, involving many language pairs that have not been studied thus far owing to the …
Microsoft Research Asia（微软亚洲研究院）；Peking University（北京大学）
In order to reveal the rationale behind model predictions, many works have exploited providing explanations in various forms. Recently, to further guarantee readability, more and more works turn to generate sentence-level human language explanations. However, current works pursuing sentence-level explanations rely heavily on annotated training data, which limits the development of interpretability to only a few tasks. As far as we know, this paper is the first to explore this problem smoothly from weak-supervised learning to unsupervised learning. Besides, we also notice the high latency of au…
In this paper, we conduct a sentence level sentiment analysis on the product reviews from Amazon and thorough analysis on the model interpretability. For the sentiment analysis task, we use the BiLSTM model with attention mechanism. For the study of interpretability, we consider the attention weights distribution of single sentence and the attention weights of main aspect terms. The model has an accuracy of up to 0.96. And we find that the aspect terms have the same or even more attention weights than the sentimental words in sentences.
Reasoning-Oriented and Analogy-Based Methods for Locating and Editing in Zero-Shot Event-Relational Reasoning
School of Computer Science and Technology, Dalian University of Technology（大连理工大学计算机科学与技术学院）；School of Computer Science and Technology, Beijing Institute of Technology（北京理工大学计算机科学与技术学院）；School of Computer Science and Artificial Intelligence, Liaoning Normal University（辽宁师范大学计算机与人工智能学院）
本文提出 ROLE 和 ABLE 两种定位与编辑方法，分别通过定位语言模型关键模块和利用任务间类比关系，高效提升零样本事件关系推理的可解释性与性能。
Tsinghua University（清华大学）；Beijing National Research Center for Information Science and Technology（北京信息科学与技术国家研究中心）；Sogou Inc.（搜狗公司）；Beijing Advanced Innovation Center for Language Resources（北京语言资源高精尖创新中心）
Automatic post-editing (APE), which aims to correct errors in the output of machine translation systems in a post-processing step, is an important task in natural language processing. While recent work has achieved considerable performance gains by using neural networks, how to model the copying mechanism for APE remains a challenge. In this work, we propose a new method for modeling copying for APE. To better identify translation errors, our method learns the representations of source sentences and system outputs in an interactive way. These representations are used to explicitly indicate whi…
