Technology for open-ended language generation, a key application of artificial intelligence, has advanced to a great extent in recent years. Large-scale language models, which are trained on large corpora of text, are being used in a wide range of applications everywhere, from virtual assistants to conversational bots. While these language models output fluent text, existing research shows that these models can and do capture human biases. Many of these biases, especially those that could potentially cause harm, are being well-investigated. On the other hand, studies that infer and change human personality traits inherited by these models have been scarce or non-existent. Our work seeks to address this gap by exploring the personality traits of several large-scale language models designed for open-ended text generation and the datasets used for training them. We build on the popular Big Five factors and develop robust methods that quantify the personality traits of these models and their underlying datasets. In particular, we trigger the models with a questionnaire designed for personality assessment and subsequently classify the text responses into quantifiable traits using a Zero-shot classifier. Our estimation scheme sheds light on an important anthropomorphic element found in such AI models and can help stakeholders decide how they should be applied as well as how society could perceive them. Additionally, we examined approaches to alter these personalities, adding to our understanding of how AI models can be adapted to specific contexts.
Language models generate texts by successively predicting probability distributions for next tokens given past ones. A growing field of interest tries to leverage external information in the decoding process so that the generated texts have desired properties, such as being more natural, non toxic, faithful, or having a specific writing style. A solution is to use a classifier at each generation step, resulting in a cooperative environment where the classifier guides the decoding of the language model distribution towards relevant texts for the task at hand. In this paper, we examine three families of (transformer-based) discriminators for this specific task of cooperative decoding: bidirectional, left-to-right and generative ones. We evaluate the pros and cons of these different types of discriminators for cooperative generation, exploring respective accuracy on classification tasks along with their impact on the resulting sample quality and computational performances. We also provide the code of a batched implementation of the powerful cooperative decoding strategy used for our experiments, the Monte Carlo Tree Search, working with each discriminator for Natural Language Generation.
Model-based, reference-free evaluation metrics have been proposed as a fast and cost-effective approach to evaluate Natural Language Generation (NLG) systems. Despite promising recent results, we find evidence that reference-free evaluation metrics of summarization and dialog generation may be relying on spurious correlations with measures such as word overlap, perplexity, and length. We further observe that for text summarization, these metrics have high error rates when ranking current state-of-the-art abstractive summarization systems. We demonstrate that these errors can be mitigated by explicitly designing evaluation metrics to avoid spurious features in reference-free evaluation.
Goal-oriented dialogue systems face a trade-off between fluent language generation and task-specific control. While supervised learning with large language models is capable of producing realistic text, how to steer such responses towards completing a specific task without sacrificing language quality remains an open question. In this work, we formulate goal-oriented dialogue as a partially observed Markov decision process, interpreting the language model as a representation of both the dynamics and the policy. This view allows us to extend techniques from learning-based control, such as task relabeling, to derive a simple and effective method to finetune language models in a goal-aware way, leading to significantly improved task performance. We additionally introduce a number of training strategies that serve to better focus the model on the task at hand. We evaluate our method, Context-Aware Language Models (CALM), on a practical flight-booking task using AirDialogue. Empirically, CALM outperforms the state-of-the-art method by 7% in terms of task success, matching human-level task performance.
This paper aims to explore a generative approach for knowledge-based design ideation by applying the latest pre-trained language models in artificial intelligence (AI). Specifically, a method of fine-tuning the generative pre-trained transformer using the USPTO patent database is proposed. The AI-generated ideas are not only in concise and understandable language but also able to synthesize the target design with external knowledge sources with controllable knowledge distance. The method is tested in a case study of rolling toy design and the results show good performance in generating ideas of varied novelty with near-field and far-field source knowledge.
Open-ended text generation tasks, such as dialogue generation and story completion, require models to generate a coherent continuation given limited preceding context. The open-ended nature of these tasks brings new challenges to the neural auto-regressive text generators nowadays. Despite these neural models are good at producing human-like text, it is difficult for them to arrange causalities and relations between given facts and possible ensuing events. To bridge this gap, we propose a novel two-stage method which explicitly arranges the ensuing events in open-ended text generation. Our approach can be understood as a specially-trained coarse-to-fine algorithm, where an event transition planner provides a "coarse" plot skeleton and a text generator in the second stage refines the skeleton. Experiments on two open-ended text generation tasks demonstrate that our proposed method effectively improves the quality of the generated text, especially in coherence and diversity. The code is available at: \url{https://github.com/qtli/EventPlanforTextGen}.
Dialogue generation is an important NLP task fraught with many challenges. The challenges become more daunting for low-resource African languages. To enable the creation of dialogue agents for African languages, we contribute the first high-quality dialogue datasets for 6 African languages: Swahili, Wolof, Hausa, Nigerian Pidgin English, Kinyarwanda & Yor\`ub\'a. These datasets consist of 1,500 turns each, which we translate from a portion of the English multi-domain MultiWOZ dataset. Subsequently, we investigate & analyze the effectiveness of modelling through transfer learning by utilziing state-of-the-art (SoTA) deep monolingual models: DialoGPT and BlenderBot. We compare the models with a simple seq2seq baseline using perplexity. Besides this, we conduct human evaluation of single-turn conversations by using majority votes and measure inter-annotator agreement (IAA). We find that the hypothesis that deep monolingual models learn some abstractions that generalize across languages holds. We observe human-like conversations, to different degrees, in 5 out of the 6 languages. The language with the most transferable properties is the Nigerian Pidgin English, with a human-likeness score of 78.1%, of which 34.4% are unanimous. We freely provide the datasets and host the model checkpoints/demos on the HuggingFace hub for public access.
Clarifying the underlying user information need by asking clarifying questions is an important feature of modern conversational search system. However, evaluation of such systems through answering prompted clarifying questions requires significant human effort, which can be time-consuming and expensive. In this paper, we propose a conversational User Simulator, called USi, for automatic evaluation of such conversational search systems. Given a description of an information need, USi is capable of automatically answering clarifying questions about the topic throughout the search session. Through a set of experiments, including automated natural language generation metrics and crowdsourcing studies, we show that responses generated by USi are both inline with the underlying information need and comparable to human-generated answers. Moreover, we make the first steps towards multi-turn interactions, where conversational search systems asks multiple questions to the (simulated) user with a goal of clarifying the user need. To this end, we expand on currently available datasets for studying clarifying questions, i.e., Qulac and ClariQ, by performing a crowdsourcing-based multi-turn data acquisition. We show that our generative, GPT2-based model, is capable of providing accurate and natural answers to unseen clarifying questions in the single-turn setting and discuss capabilities of our model in the multi-turn setting. We provide the code, data, and the pre-trained model to be used for further research on the topic.
For multilingual sequence-to-sequence pretrained language models (multilingual Seq2Seq PLMs), e.g. mBART, the self-supervised pretraining task is trained on a wide range of monolingual languages, e.g. 25 languages from CommonCrawl, while the downstream cross-lingual tasks generally progress on a bilingual language subset, e.g. English-German, making there exists the data discrepancy, namely domain discrepancy, and cross-lingual learning objective discrepancy, namely task discrepancy, between the pretraining and finetuning stages. To bridge the above cross-lingual domain and task gaps, we extend the vanilla pretrain-finetune pipeline with extra code-switching restore task. Specifically, the first stage employs the self-supervised code-switching restore task as a pretext task, allowing the multilingual Seq2Seq PLMs to acquire some in-domain alignment information. And for the second stage, we fine-tune the model on downstream data normally. Experiments on both NLG evaluation (12 bilingual translation tasks, 30 zero-shot translation tasks, and 2 cross-lingual summarization tasks) and NLU evaluation (7 cross-lingual natural language inference tasks) show our model outperforms the strong baseline mBART with standard finetuning strategy, consistently. Analyses indicate our approach could narrow the Euclidean distance of cross-lingual sentence representations, and improve the model generalization with trivial computational cost. We release the code at: https://github.com/zanchangtong/CSR4mBART.
Controlled text generation tasks such as unsupervised text style transfer have increasingly adopted the use of Reinforcement Learning (RL). A major challenge in applying RL to such tasks is the sparse reward, which is available only after the full text is generated. Sparse rewards, combined with a large action space make RL training sample-inefficient and difficult to converge. Recently proposed reward-shaping strategies to address this issue have shown only negligible gains. In contrast, this work proposes a novel approach that provides dense rewards to each generated token. We evaluate our approach by its usage in unsupervised text style transfer. Averaged across datasets, our style transfer system improves upon current state-of-art systems by 21\% on human evaluation and 12\% on automatic evaluation. Upon ablated comparison with the current reward shaping approach (the `roll-out strategy'), using dense rewards improves the overall style transfer quality by 22\% based on human evaluation. Further the RL training is 2.5 times as sample efficient, and 7 times faster.
Word ordering is a constrained language generation task taking unordered words as input. Existing work uses linear models and neural networks for the task, yet pre-trained language models have not been studied in word ordering, let alone why they help. We use BART as an instance and show its effectiveness in the task. To explain why BART helps word ordering, we extend analysis with probing and empirically identify that syntactic dependency knowledge in BART is a reliable explanation. We also report performance gains with BART in the related partial tree linearization task, which readily extends our analysis.
Text revision refers to a family of natural language generation tasks, where the source and target sequences share moderate resemblance in surface form but differentiate in attributes, such as text formality and simplicity. Current state-of-the-art methods formulate these tasks as sequence-to-sequence learning problems, which rely on large-scale parallel training corpus. In this paper, we present an iterative in-place editing approach for text revision, which requires no parallel data. In this approach, we simply fine-tune a pre-trained Transformer with masked language modeling and attribute classification. During inference, the editing at each iteration is realized by two-step span replacement. At the first step, the distributed representation of the text optimizes on the fly towards an attribute function. At the second step, a text span is masked and another new one is proposed conditioned on the optimized representation. The empirical experiments on two typical and important text revision tasks, text formalization and text simplification, show the effectiveness of our approach. It achieves competitive and even better performance than state-of-the-art supervised methods on text simplification, and gains better performance than strong unsupervised methods on text formalization \footnote{Code and model are available at \url{https://github.com/jingjingli01/OREO}}.
Recent improvements in KG-to-text generation are due to additional auxiliary pre-training tasks designed to give the fine-tune task a boost in performance. These tasks require extensive computational resources while only suggesting marginal improvements. Here, we demonstrate that by fusing graph-aware elements into existing pre-trained language models, we are able to outperform state-of-the-art models and close the gap imposed by additional pre-training tasks. We do so by proposing a mask structure to capture neighborhood information and a novel type encoder that adds a bias to the graph-attention weights depending on the connection type. Experiments on two KG-to-text benchmark datasets show our models are competitive while involving fewer parameters and no additional pre-training tasks. By formulating the problem as a framework, we can interchange the various proposed components and begin interpreting KG-to-text generative models based on the topological and type information found in a graph.
Row completion is the task of augmenting a given table of text and numbers with additional, relevant rows. The task divides into two steps: subject suggestion, the task of populating the main column; and gap filling, the task of populating the remaining columns. We present state-of-the-art results for subject suggestion and gap filling measured on a standard benchmark (WikiTables). Our idea is to solve this task by harmoniously combining knowledge base table interpretation and free text generation. We interpret the table using the knowledge base to suggest new rows and generate metadata like headers through property linking. To improve candidate diversity, we synthesize additional rows using free text generation via GPT-3, and crucially, we exploit the metadata we interpret to produce better prompts for text generation. Finally, we verify that the additional synthesized content can be linked to the knowledge base or a trusted web source such as Wikipedia.
Retrosynthesis analysis is a critical task in organic chemistry central to many important industries. Previously, various machine learning approaches have achieved promising results on this task by representing output molecules as strings and autoregressively decoded token-by-token with generative models. Text generation or machine translation models in natural language processing were frequently utilized approaches. The token-by-token decoding approach is not intuitive from a chemistry perspective because some substructures are relatively stable and remain unchanged during reactions. In this paper, we propose a substructure-level decoding model, where the substructures are reaction-aware and can be automatically extracted with a fully data-driven approach. Our approach achieved improvement over previously reported models, and we find that the performance can be further boosted if the accuracy of substructure extraction is improved. The substructures extracted by our approach can provide users with better insights for decision-making compared to existing methods. We hope this work will generate interest in this fast growing and highly interdisciplinary area on retrosynthesis prediction and other related topics.
Text generation with beam search has proven successful in a wide range of applications. We point out that, though largely overlooked in the literature, the commonly-used implementation of beam decoding (e.g., Hugging Face Transformers and fairseq) uses a first come, first served heuristic: it keeps a set of already completed sequences over time steps and stops when the size of this set reaches the beam size. Based on this finding, we introduce a patience factor, a simple modification to this beam decoding implementation, that generalizes the stopping criterion and provides flexibility to the depth of search. Empirical results demonstrate that adjusting this patience factor improves decoding performance of strong pretrained models on news text summarization and machine translation over diverse language pairs, with a negligible inference slowdown. Our approach only modifies one line of code and can be thus readily incorporated in any implementation. Further, we find that different versions of beam decoding result in large performance differences in summarization, demonstrating the need for clarity in specifying the beam search implementation in research work. Our code will be available upon publication.
Sentiment analysis is an important task in natural language processing. In recent works, pre-trained language models are often used to achieve state-of-the-art results, especially when training data is scarce. It is common to fine-tune on the downstream task, usually by adding task-specific layers on top of the model. In this paper, we focus on aspect-based sentiment analysis, which involves extracting aspect term, category, and predicting their corresponding polarities. In particular, we are interested in few-shot settings. We propose to reformulate the extraction and prediction tasks into the sequence generation task, using a generative language model with unidirectional attention (GPT2 is used unless stated otherwise). This way, the model learns to accomplish the tasks via language generation without the need of training task-specific layers. Our evaluation results on the single-task polarity prediction show that our approach outperforms the previous state-of-the-art (based on BERT) on average performance by a large margins in few-shot and full-shot settings. More importantly, our generative approach significantly reduces the model variance caused by low-resource data. We further demonstrate that the proposed generative language model can handle joint and multi-task settings, unlike previous work. We observe that the proposed sequence generation method achieves further improved performances on polarity prediction when the model is trained via joint and multi-task settings. Further evaluation on similar sentiment analysis datasets, SST-2, SST- and OOS intent detection validates the superiority and noise robustness of generative language model in few-shot settings.
Grounded text generation systems often generate text that contains factual inconsistencies, hindering their real-world applicability. Automatic factual consistency evaluation may help alleviate this limitation by accelerating evaluation cycles, filtering inconsistent outputs and augmenting training data. While attracting increasing attention, such evaluation metrics are usually developed and evaluated in silo for a single task or dataset, slowing their adoption. Moreover, previous meta-evaluation protocols focused on system-level correlations with human annotations, which leave the example-level accuracy of such metrics unclear. In this work, we introduce TRUE: a comprehensive survey and assessment of factual consistency metrics on a standardized collection of existing texts from diverse tasks, manually annotated for factual consistency. Our standardization enables an example-level meta-evaluation protocol that is more actionable and interpretable than previously reported correlations, yielding clearer quality measures. Across diverse state-of-the-art metrics and 11 datasets we find that large-scale NLI and question generation-and-answering-based approaches achieve strong and complementary results. We recommend those methods as a starting point for model and metric developers, and hope TRUE will foster progress towards even better evaluation methods.
Improvements in text generation technologies such as machine translation have necessitated more costly and time-consuming human evaluation procedures to ensure an accurate signal. We investigate a simple way to reduce cost by reducing the number of text segments that must be annotated in order to accurately predict a score for a complete test set. Using a sampling approach, we demonstrate that information from document membership and automatic metrics can help improve estimates compared to a pure random sampling baseline. We achieve gains of up to 20% in average absolute error by leveraging stratified sampling and control variates. Our techniques can improve estimates made from a fixed annotation budget, are easy to implement, and can be applied to any problem with structure similar to the one we study.
Large language models (LLMs) have shown promising results in a wide array of generative NLP tasks, such as summarization and machine translation. In the context of narrative generation, however, existing models still do not capture factors that contribute to producing consistent text. For instance, it is logical that a piece of text or a story should be uniformly readable throughout and that this form of complexity should be controllable. As such, if the complexity of an input text prompt is rated first-grade reading level in the Flesch Reading Ease test, then the generated text continuing the plot should also be within this range of complexity. With this in mind, we introduce Uniform Complexity for Text Generation (UCTG), a new benchmark test which raises the challenge of making generative models observe uniform linguistic properties with respect to prompts. We experiment with over 150+ linguistically and cognitively motivated features for evaluating text complexity in humans and generative models. From our results, we find that models such as GPT-2 struggle to preserve the complexity of input prompts used in its generations, even if finetuned with professionally written texts.