Learners that are exposed to the same training data might generalize differently due to differing inductive biases. In neural network models, inductive biases could in theory arise from any aspect of the model architecture. We investigate which architectural factors affect the generalization behavior of neural sequence-to-sequence models trained on two syntactic tasks, English question formation and English tense reinflection. For both tasks, the training set is consistent with a generalization based on hierarchical structure and a generalization based on linear order. All architectural factors that we investigated qualitatively affected how models generalized, including factors with no clear connection to hierarchical structure. For example, LSTMs and GRUs displayed qualitatively different inductive biases. However, the only factor that consistently contributed a hierarchical bias across tasks was the use of a tree-structured model rather than a model with sequential recurrence, suggesting that human-like syntactic generalization requires architectural syntactic structure.
LIPN（巴黎第十三大学信息与通信科学实验室）；Université Sorbonne Paris Nord（巴黎第十三大学）；CNRS（法国国家科学研究中心）；INRIA Paris（巴黎法国国家信息与自动化研究所）
We propose a generative model for text generation, which exhibits disentangled latent representations of syntax and semantics. Contrary to previous work, this model does not need syntactic information such as constituency parses, or semantic information such as paraphrase pairs. Our model relies solely on the inductive bias found in attention-based architectures such as Transformers. In the attention of Transformers, keys handle information selection while values specify what information is conveyed. Our model, dubbed QKVAE, uses Attention in its decoder to read latent variables where one late…
Lucy Software（露西软件公司）；United Language Group（联合语言集团）；TALP Research Center, Universitat Politècnica de Catalunya（加泰罗尼亚理工大学TALP研究中心）
The dominant language modeling paradigm handles text as a sequence of discrete tokens. While that approach can capture the latent structure of the text, it is inherently constrained to sequential dynamics for text generation. We propose a new paradigm for introducing a syntactic inductive bias into neural text generation, where the dependency parse tree is used to drive the Transformer model to generate sentences iteratively. Our experiments show that this paradigm is effective at text generation, with quality between LSTMs and Transformers, and comparable diversity, requiring less than half t…
We introduce Transformer Grammars (TGs), a novel class of Transformer language models that combine (i) the expressive power, scalability, and strong performance of Transformers and (ii) recursive syntactic compositions, which here are implemented through a special attention mask and deterministic transformation of the linearized tree. We find that TGs outperform various strong baselines on sentence-level language modeling perplexity, as well as on multiple syntax-sensitive language modeling evaluation metrics. Additionally, we find that the recursive syntactic composition bottleneck which repr…
Bar Ilan University（巴伊兰大学）；Allen Institute for Artificial Intelligence（艾伦人工智能研究所）；Johns Hopkins University（约翰斯·霍普金斯大学）
How do typological properties such as word order and morphological case marking affect the ability of neural sequence models to acquire the syntax of a language? Cross-linguistic comparisons of RNNs' syntactic performance (e.g., on subject-verb agreement prediction) are complicated by the fact that any two languages differ in multiple typological properties, as well as by differences in training corpus. We propose a paradigm that addresses these issues: we create synthetic versions of English, which differ from English in one or more typological parameters, and generate corpora for those langu…
Long short-term memory (LSTM) networks and their variants are capable of encapsulating long-range dependencies, which is evident from their performance on a variety of linguistic tasks. On the other hand, simple recurrent networks (SRNs), which appear more biologically grounded in terms of synaptic connections, have generally been less successful at capturing long-range dependencies as well as the loci of grammatical errors in an unsupervised setting. In this paper, we seek to develop models that bridge the gap between biological plausibility and linguistic competence. We propose a new archite…
In the last half-decade, the field of natural language processing (NLP) has undergone two major transitions: the switch to neural networks as the primary modeling paradigm and the homogenization of the training regime (pre-train, then fine-tune). Amidst this process, language models have emerged as NLP's workhorse, displaying increasingly fluent generation capabilities and proving to be an indispensable means of knowledge transfer downstream. Due to the otherwise opaque, black-box nature of such models, researchers have employed aspects of linguistic theory in order to characterize their behav…
Children learning their first language face multiple problems of induction: how to learn the meanings of words, and how to build meaningful phrases from those words according to syntactic rules. We consider how children might solve these problems efficiently by solving them jointly, via a computational model that learns the syntax and semantics of multi-word utterances in a grounded reference game. We select a well-studied empirical case in which children are aware of patterns linking the syntactic and semantic properties of words --- that the properties picked out by base nouns tend to be rel…
Allen Institute for AI（艾伦人工智能研究所）；New York University（纽约大学）；University of Washington（华盛顿大学）；Bar Ilan University（巴伊兰大学）；Hebrew University of Jerusalem（希伯来大学）
The capacity of neural networks like the widely adopted transformer is known to be very high. Evidence is emerging that they learn successfully due to inductive bias in the training routine, typically a variant of gradient descent (GD). To better understand this bias, we study the tendency for transformer parameters to grow in magnitude ($\ell_2$ norm) during training, and its implications for the emergent representations within self attention layers. Empirically, we document norm growth in the training of transformer language models, including T5 during its pretraining. As the parameters grow…
Florida State University（佛罗里达州立大学）；Tencent AI Lab（腾讯人工智能实验室）
Recent studies have shown that a hybrid of self-attention networks (SANs) and recurrent neural networks (RNNs) outperforms both individual architectures, while not much is known about why the hybrid models work. With the belief that modeling hierarchical structure is an essential complementary between SANs and RNNs, we propose to further enhance the strength of hybrid models with an advanced variant of RNNs - Ordered Neurons LSTM (ON-LSTM), which introduces a syntax-oriented inductive bias to perform tree-like composition. Experimental results on the benchmark machine translation task show tha…
Recent work in NLP shows that LSTM language models capture hierarchical structure in language data. In contrast to existing work, we consider the \textit{learning} process that leads to their compositional behavior. For a closer look at how an LSTM's sequential representations are composed hierarchically, we present a related measure of Decompositional Interdependence (DI) between word meanings in an LSTM, based on their gate interactions. We connect this measure to syntax with experiments on English language data, where DI is higher on pairs of words with lower syntactic distance. To explore …
Dependency Transformer Grammars: Integrating Dependency Structures into Transformer Language Models
School of Information Science and Technology, ShanghaiTech University（上海科技大学信息科学与技术学院）；Shanghai Engineering Research Center of Intelligent Vision and Imaging（上海智能视觉与成像工程技术研究中心）
Do Robot Snakes Dream like Electric Sheep? Investigating the Effects of Architectural Inductive Biases on Hallucination
Mila & Université de Montréal（Mila与蒙特利尔大学）；Noah’s Ark Lab（Noah’s Ark实验室）；Advanced Micro Devices；Chandar Research Lab（Chandar研究实验室）；Polytechnique Montréal（蒙特利尔理工学院）；CIFAR AI Chair（CIFAR人工智能主席）
