Pretrained language models have been shown to significantly predict brain recordings of people comprehending language. Recent work suggests that the prediction of the next word is a key mechanism that contributes to this alignment. What is not yet understood is whether prediction of the next word is necessary for this observed alignment or simply sufficient, and whether there are other shared mechanisms or information that are similarly important. In this work, we take a step towards understanding the reasons for brain alignment via two simple perturbations in popular pretrained language models. These perturbations help us design contrasts that can control for different types of information. By contrasting the brain alignment of these differently perturbed models, we show that improvements in alignment with brain recordings are due to more than improvements in next-word prediction and word-level information.
Explanations of Large Language Models Explain Language Representations in the Brain
Iran University of Science and Technology（伊朗科学技术大学）；University of Tehran（德黑兰大学）；Institute for Research in Fundamental Sciences（基础科学研究所）；Khatam University（哈塔姆大学）
Improve Language Model and Brain Alignment via Associative Memory
College of Artificial Intelligence, Nanjing University of Aeronautics and Astronautics（人工智能学院，南京航空航天大学）；The Key Laboratory of Brain-Machine Intelligence Technology, Ministry of Education（脑机智能技术重点实验室，教育部）
Large Language Models Align with the Human Brain during Creative Thinking
EPFL（瑞士联邦理工学院洛桑）；Università della Svizzera italiana (USI)（意大利语区大学（瑞士））；Wesleyan University（卫斯理大学）；Paris Brain Institute (ICM)（巴黎大脑研究所）；Pennsylvania State University（宾夕法尼亚州立大学）
Beyond representational alignment with brain-guided language models for robust reasoning
State Key Lab of General AI, School of Intelligence Science and Technology, Peking University（北京大学通用人工智能国家重点实验室、智能科学与技术学院）；Department of Psychological and Cognitive Sciences, Tsinghua University（清华大学心理与认知科学系）；Microsoft Research Asia（微软亚洲研究院）
Measuring Meaning Composition in the Human Brain with Composition Scores from Large Language Models
National Key Laboratory for Novel Software Technology, Nanjing University（南京大学计算机软件新技术国家重点实验室）；Department of Linguistics and Translation, City University of Hong Kong（香港城市大学语言及翻译学系）
Decoding the Multimodal Mind: Generalizable Brain-to-Text Translation via Multimodal Alignment and Adaptive Routing
State Key Laboratory of Multimodal Artificial Intelligence System, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Department of Computer Science, The University of Manchester（曼彻斯特大学计算机科学系）；Department of Language Science and Technology, Hong Kong Polytechnic University（香港理工大学语言科学与技术系）
Do Large Language Models Mirror Cognitive Language Processing?
College of Intelligence and Computing, Tianjin University（天津大学智能与计算学部）；College of Computer and Information Engineering, Tianjin Normal University（天津师范大学计算机与信息工程学院）
Shanghai University of Finance and Economics（上海财经大学）；Southern University of Science and Technology（南方科技大学）
Multilingual pretrained language models (mPLMs) have shown their effectiveness in multilingual word alignment induction. However, these methods usually start from mBERT or XLM-R. In this paper, we investigate whether multilingual sentence Transformer LaBSE is a strong multilingual word aligner. This idea is non-trivial as LaBSE is trained to learn language-agnostic sentence-level embeddings, while the alignment extraction task requires the more fine-grained word-level embeddings to be language-agnostic. We demonstrate that the vanilla LaBSE outperforms other mPLMs currently used in the alignme…
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络空间安全学院）
Most Chinese pre-trained models take character as the basic unit and learn representation according to character's external contexts, ignoring the semantics expressed in the word, which is the smallest meaningful utterance in Chinese. Hence, we propose a novel word-aligned attention to exploit explicit word information, which is complementary to various character-based Chinese pre-trained language models. Specifically, we devise a pooling mechanism to align the character-level attention to the word level and propose to alleviate the potential issue of segmentation error propagation by multi-so…
Accurate Decoding of Natural Sentences from Non-Invasive Brain Recordings
Meta AI；École Normale Supérieure（高等师范学院）；Université PSL（巴黎文理大学）；CNRS（法国国家科学研究中心）；Hospital Foundation Adolphe de Rothschild（阿道夫·德·罗斯柴尔德医院基金会）；Univ. Lille（里尔大学）；Basque Center on Cognition, Brain and Language（巴斯克认知、大脑与语言中心）；Paris Cité University（巴黎城市大学）；Inria（法国国家信息与自动化研究所）；Université Paris-Saclay（巴黎萨克雷大学）；INSERM（法国国家健康与医学研究院）；CEA（法国原子能和替代能源委员会）
We evaluate 8 different word embedding models on their usefulness for predicting the neural activation patterns associated with concrete nouns. The models we consider include an experiential model, based on crowd-sourced association data, several popular neural and distributional models, and a model that reflects the syntactic context of words (based on dependency parses). Our goal is to assess the cognitive plausibility of these various embedding models, and understand how we can further improve our methods for interpreting brain imaging data. We show that neural word embedding models exhibit…
Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology（麻省理工学院脑与认知科学系）
Language decoding studies have identified word representations which can be used to predict brain activity in response to novel words and sentences (Anderson et al., 2016; Pereira et al., 2018). The unspoken assumption of these studies is that, during processing, linguistic information is transformed into some shared semantic space, and those semantic representations are then used for a variety of linguistic and non-linguistic tasks. We claim that current studies vastly underdetermine the content of these representations, the algorithms which the brain deploys to produce and consume them, and …
Brain decoding, understood as the process of mapping brain activities to the stimuli that generated them, has been an active research area in the last years. In the case of language stimuli, recent studies have shown that it is possible to decode fMRI scans into an embedding of the word a subject is reading. However, such word embeddings are designed for natural language processing tasks rather than for brain decoding. Therefore, they limit our ability to recover the precise stimulus. In this work, we propose to directly classify an fMRI scan, mapping it to the corresponding word within a fixe…
The pervasive use of distributional semantic models or word embeddings in a variety of research fields is due to their remarkable ability to represent the meanings of words for both practical application and cognitive modeling. However, little has been known about what kind of information is encoded in text-based word vectors. This lack of understanding is particularly problematic when word vectors are regarded as a model of semantic representation for abstract concepts. This paper attempts to reveal the internal information of distributional word vectors by the analysis using Binder et al.'s …
University of Oxford（牛津大学）；Max Planck Institute for Psycholinguistics（马克斯·普朗克心理语言学研究所）；Radboud University（拉德堡德大学）
Recent artificial neural networks that process natural language achieve unprecedented performance in tasks requiring sentence-level understanding. As such, they could be interesting models of the integration of linguistic information in the human brain. We review works that compare these artificial language models with human brain activity and we assess the extent to which this approach has improved our understanding of the neural processes involved in natural language comprehension. Two main results emerge. First, the neural representation of word meaning aligns with the context-dependent, de…
Max Planck Institute for Software Systems（马克斯·普朗克软件系统研究所）；Singapore Management University（新加坡管理大学）
Building systems that achieve a deeper understanding of language is one of the central goals of natural language processing (NLP). Towards this goal, recent works have begun to train language models on narrative datasets which require extracting the most critical information by integrating across long contexts. However, it is still an open question whether these models are learning a deeper understanding of the text, or if the models are simply learning a heuristic to complete the task. This work investigates this further by turning to the one language processing system that truly understands …
ITESM（蒙特雷理工学院）；The University of Texas at Austin（德克萨斯大学奥斯汀分校）
Semantic feature models have become a popular tool for prediction and interpretation of fMRI data. In particular, prior work has shown that differences in the fMRI patterns in sentence reading can be explained by context-dependent changes in the semantic feature representations of the words. However, whether the subjects are aware of such changes and agree with them has been an open question. This paper aims to answer this question through a human-subject study. Subjects were asked to judge how the word change from their generic meaning when the words were used in specific sentences. The judge…
This article presents a neural language architecture of sentence structure in the brain, in which sentences are temporal connection paths that interconnect neural structures underlying their words. Words remain 'in-situ', hence they are always content-addressable. Arbitrary and novel sentences (with novel words) can be created with 'neural blackboards' for words and sentences. Hence, the unlimited productivity of natural language can be achieved with a 'fixed' small world like network structure. The article focuses on the neural blackboard for sentences. The architecture uses only one 'connect…
How the brain captures the meaning of linguistic stimuli across multiple views is still a critical open question in neuroscience. Consider three different views of the concept apartment: (1) picture (WP) presented with the target word label, (2) sentence (S) using the target word, and (3) word cloud (WC) containing the target word along with other semantically related words. Unlike previous efforts, which focus only on single view analysis, in this paper, we study the effectiveness of brain decoding in a zero-shot cross-view learning setup. Further, we propose brain decoding in the novel conte…
The dispute of how the human brain represents conceptual knowledge has been argued in many scientific fields. Brain imaging studies have shown that the spatial patterns of neural activation in the brain are correlated with thinking about different semantic categories of words (for example, tools, animals, and buildings) or when viewing the related pictures. In this paper, we present a computational model that learns to predict the neural activation captured in functional magnetic resonance imaging (fMRI) data of test words. Unlike the models with hand-crafted features that have been used in th…
Word alignment is to find translationally equivalent words between source and target sentences. Previous work has demonstrated that self-training can achieve competitive word alignment results. In this paper, we propose to use word alignments generated by a third-party word aligner to supervise the neural word alignment training. Specifically, source word and target word of each word pair aligned by the third-party aligner are trained to be close neighbors to each other in the contextualized embedding space when fine-tuning a pre-trained cross-lingual language model. Experiments on the benchma…
How meaning is represented in the brain is still one of the big open questions in neuroscience. Does a word (e.g., bird) always have the same representation, or does the task under which the word is processed alter its representation (answering "can you eat it?" versus "can it fly?")? The brain activity of subjects who read the same word while performing different semantic tasks has been shown to differ across tasks. However, it is still not understood how the task itself contributes to this difference. In the current work, we study Magnetoencephalography (MEG) brain recordings of participants…
We propose a variation to the commonly used Word Error Rate (WER) metric for speech recognition evaluation which incorporates the alignment of phonemes, in the absence of time boundary information. After computing the Levenshtein alignment on words in the reference and hypothesis transcripts, spans of adjacent errors are converted into phonemes with word and syllable boundaries and a phonetic Levenshtein alignment is performed. The aligned phonemes are recombined into aligned words that adjust the word alignment labels in each error region. We demonstrate that our Phonetically-Oriented Word Er…
The word alignment task, despite its prominence in the era of statistical machine translation (SMT), is niche and under-explored today. In this two-part tutorial, we argue for the continued relevance for word alignment. The first part provides a historical background to word alignment as a core component of the traditional SMT pipeline. We zero-in on GIZA++, an unsupervised, statistical word aligner with surprising longevity. Jumping forward to the era of neural machine translation (NMT), we show how insights from word alignment inspired the attention mechanism fundamental to present-day NMT. …
Instituto Nacional de Astrofísica, Óptica y Electrónica（国家天体物理、光学与电子研究所）
A Brain Computer Interface based on imagined words can decode the word a subject is thinking on through brain signals to control an external device. In order to build a fully asynchronous Brain Computer Interface based on imagined words in electroencephalogram signals as source, we need to solve the problem of detecting the onset of the imagined words. Although there has been some research in this field, the problem has not been fully solved. In this paper we present an approach to solve this problem by using values from statistics, information theory and chaos theory as features to correctly …
Idiap Research Institute（伊迪亚普研究所）；Ecole Polytechnique Fédérale de Lausanne (EPFL)（洛桑联邦理工学院）；Facebook AI Research（脸书人工智能研究院）
We present a simple neural network for word alignment that builds source and target word window representations to compute alignment scores for sentence pairs. To enable unsupervised training, we use an aggregation operation that summarizes the alignment scores for a given target word. A soft-margin objective increases scores for true target words while decreasing scores for target words that are not present. Compared to the popular Fast Align model, our approach improves alignment accuracy by 7 AER on English-Czech, by 6 AER on Romanian-English and by 1.7 AER on English-French alignment.
Nara Institute of Science and Technology（奈良先端科学技术大学）；Tokyo Metropolitan University（东京都立大学）
One of the most important problems in machine translation (MT) evaluation is to evaluate the similarity between translation hypotheses with different surface forms from the reference, especially at the segment level. We propose to use word embeddings to perform word alignment for segment-level MT evaluation. We performed experiments with three types of alignment methods using word embeddings. We evaluated our proposed methods with various translation datasets. Experimental results show that our proposed methods outperform previous word embeddings-based methods.
Ritsumeikan University（立命馆大学）；The Whole Brain Architecture Initiative（全脑架构倡议组织）；The University of Tokyo（东京大学）；RIKEN（理化学研究所）
The human brain, among its several functions, analyzes the double articulation structure in spoken language, i.e., double articulation analysis (DAA). A hierarchical structure in which words are connected to form a sentence and words are composed of phonemes or syllables is called a double articulation structure. Where and how DAA is performed in the human brain has not been established, although some insights have been obtained. In addition, existing computational models based on a probabilistic generative model (PGM) do not incorporate neuroscientific findings, and their consistency with the…
Word alignments are useful for tasks like statistical and neural machine translation (NMT) and cross-lingual annotation projection. Statistical word aligners perform well, as do methods that extract alignments jointly with translations in NMT. However, most approaches require parallel training data, and quality decreases as less training data is available. We propose word alignment methods that require no parallel data. The key idea is to leverage multilingual word embeddings, both static and contextualized, for word alignment. Our multilingual embeddings are created from monolingual data only…
Center for Information and Language Processing (CIS), LMU Munich（慕尼黑大学信息与语言处理中心）；Université Paris-Saclay（巴黎萨克雷大学）；CNRS（法国国家科学研究中心）；LISN（法国国家信息与自动化研究所跨学科数字科学实验室）
After a period of decrease, interest in word alignments is increasing again for their usefulness in domains such as typological research, cross-lingual annotation projection, and machine translation. Generally, alignment algorithms only use bitext and do not make use of the fact that many parallel corpora are multiparallel. Here, we compute high-quality word alignments between multiple language pairs by considering all language pairs together. First, we create a multiparallel word alignment graph, joining all bilingual word alignment pairs in one graph. Next, we use graph neural networks (GNNs…
Word alignment was once a core unsupervised learning task in natural language processing because of its essential role in training statistical machine translation (MT) models. Although unnecessary for training neural MT models, word alignment still plays an important role in interactive applications of neural machine translation, such as annotation transfer and lexicon injection. While statistical MT methods have been replaced by neural approaches with superior performance, the twenty-year-old GIZA++ toolkit remains a key component of state-of-the-art word alignment systems. Prior work on neur…
NTT Communication Science Laboratories, NTT Corporation（日本电信电话株式会社NTT通信科学实验室）
We present a novel supervised word alignment method based on cross-language span prediction. We first formalize a word alignment problem as a collection of independent predictions from a token in the source sentence to a span in the target sentence. As this is equivalent to a SQuAD v2.0 style question answering task, we then solve this problem by using multilingual BERT, which is fine-tuned on a manually created gold word alignment data. We greatly improved the word alignment accuracy by adding the context of the token to the question. In the experiments using five word alignment datasets amon…
Word alignment over parallel corpora has a wide variety of applications, including learning translation lexicons, cross-lingual transfer of language processing tools, and automatic evaluation or analysis of translation outputs. The great majority of past work on word alignment has worked by performing unsupervised learning on parallel texts. Recently, however, other work has demonstrated that pre-trained contextualized word embeddings derived from multilingually trained language models (LMs) prove an attractive alternative, achieving competitive results on the word alignment task even in the a…
Cross-lingual brain-language model alignment is robust but challenges hierarchical and computational accounts
Grammar and Cognition Lab, Department of Translation & Language Sciences, Universitat Pompeu Fabra（语言与翻译科学系语法与认知实验室，庞培法华大学）；Department of Adult Psychiatry and Psychotherapy, University of Zurich（苏黎世大学成人精神病学与心理治疗系）；Neuroscience Center Zurich, University of Zurich and ETH Zurich（苏黎世大学神经科学中心与苏黎世联邦理工学院）；Center for Clinical Neuroscience and Cognition and Department of Psychiatry, University of Groningen, University Medical Center Groningen（格罗宁根大学临床神经科学与认知中心及精神病学系，格罗宁根大学医学中心）；Scalable Scientific Machine Learning Lab, Imperial College London, Department of Earth Science and Engineering（伦敦帝国理工学院可扩展科学机器学习实验室，地球科学与工程系）；Institut Català de Recerca i Estudis Avançats (ICREA), Barcelona, Spain（加泰罗尼亚高级研究与研究机构（ICREA），巴塞罗那，西班牙）
Multi-layer models with multiple attention heads per layer provide superior translation quality compared to simpler and shallower models, but determining what source context is most relevant to each target word is more challenging as a result. Therefore, deriving high-accuracy word alignments from the activations of a state-of-the-art neural machine translation model is an open challenge. We propose a simple model extension to the Transformer architecture that makes use of its hidden representations and is restricted to attend solely on encoder information to predict the next word. It can be t…
Ohio State University（俄亥俄州立大学）；Georgia Institute of Technology（佐治亚理工学院）
Monolingual word alignment is important for studying fine-grained editing operations (i.e., deletion, addition, and substitution) in text-to-text generation tasks, such as paraphrase generation, text simplification, neutralizing biased language, etc. In this paper, we present a novel neural semi-Markov CRF alignment model, which unifies word and phrase alignments through variable-length spans. We also create a new benchmark with human annotations that cover four different text genres to evaluate monolingual word alignment models in more realistic settings. Experimental results show that our pr…
