Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans can generally perform a new language task from only a few examples or from simple instructions - something which current NLP systems still largely struggle to do. Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model. GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning or domain adaptation, such as unscrambling words, using a novel word in a sentence, or performing 3-digit arithmetic. At the same time, we also identify some datasets where GPT-3's few-shot learning still struggles, as well as some datasets where GPT-3 faces methodological issues related to training on large web corpora. Finally, we find that GPT-3 can generate samples of news articles which human evaluators have difficulty distinguishing from articles written by humans. We discuss broader societal impacts of this finding and of GPT-3 in general.

/pdf/language-models-are-few-shot-learners-2fg8gvia7m.pdf

Language Models are Few-Shot Learners

Transfer learning, where a model is first pre-trained on a data-rich task before being fine-tuned on a downstream task, has emerged as a powerful technique in natural language processing (NLP). The effectiveness of transfer learning has given rise to a diversity of approaches, methodology, and practice. In this paper, we explore the landscape of transfer learning techniques for NLP by introducing a unified framework that converts all text-based language problems into a text-to-text format. Our systematic study compares pre-training objectives, architectures, unlabeled data sets, transfer approaches, and other factors on dozens of language understanding tasks. By combining the insights from our exploration with scale and our new ``Colossal Clean Crawled Corpus'', we achieve state-of-the-art results on many benchmarks covering summarization, question answering, text classification, and more. To facilitate future work on transfer learning for NLP, we release our data set, pre-trained models, and code.

Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer

Part I. Experimental Studies: 2. Experiment in psychology 3. Experiments on perceiving III Experiments on imaging 4-8. Experiments on remembering: (a) The method of description (b) The method of repeated reproduction (c) The method of picture writing (d) The method of serial reproduction (e) The method of serial reproduction picture material 9. Perceiving, recognizing, remembering 10. A theory of remembering 11. Images and their functions 12. Meaning Part II. Remembering as a Study in Social Psychology: 13. Social psychology 14. Social psychology and the matter of recall 15. Social psychology and the manner of recall 16. Conventionalism 17. The notion of a collective unconscious 18. The basis of social recall 19. A summary and some conclusions.

Remembering. A Study in Experimental and Social Psychology, Cambridge (University Press) 1964.

Recent progress in natural language processing has been driven by advances in both model architecture and model pretraining. Transformer architectures have facilitated building higher-capacity models and pretraining has made it possible to effectively utilize this capacity for a wide variety of tasks. \textit{Transformers} is an open-source library with the goal of opening up these advances to the wider machine learning community. The library consists of carefully engineered state-of-the art Transformer architectures under a unified API. Backing this library is a curated collection of pretrained models made by and available for the community. \textit{Transformers} is designed to be extensible by researchers, simple for practitioners, and fast and robust in industrial deployments. The library is available at \url{this https URL}.

HuggingFace's Transformers: State-of-the-art Natural Language Processing.

A wide variety of media can be used in learning, including distance learning, such as print, lectures, conference sections, tutors, pictures, video, sound, and computers. Any one instance of distance learning will make choices among these media, perhaps using several.

Multimedia in Learning

Contents: Strategies and Cognition. Strategic Analogies. Representational Requirements. Strategy Representations. Conceptual Index to Strategies. Representational Areas.

Strategy Representation: An Analysis of Planning Knowledge

■ The central challenge in commonsense knowledge representation research is to develop content theories that achieve a high degree of both competency and coverage. We describe a new methodology for constructing formal theories in commonsense knowledge domains that complements traditional knowledge representation approaches by first addressing issues of coverage. We show how a close examination of a very general task (strategic planning) leads to a catalog of the concepts and facts that must be encoded for general commonsense reasoning. These concepts are sorted into a manageable number of coherent domains, one of which is the representational area of commonsense human memory. We then elaborate on these concepts using textual corpus-analysis techniques, where the conceptual distinctions made in natural language are used to improve the definitions of the concepts that should be expressible in our formal theories. These representational areas are then analyzed using more traditional knowledge representation techniques, as demonstrated in this article by our treatment of commonsense human memory.

/pdf/formalizations-of-commonsense-psychology-134xrjg071.pdf

Formalizations of Commonsense Psychology

The increased use of large corpora in narrative research has created new opportunities for empirical research and intelligent narrative technologies. To best exploit the value of these corpora, several research groups are eschewing complex discourse analysis techniques in favor of high-level minimalist narrative annotation schemes that can be quickly applied, achieve high inter-rater agreement, and are amenable to automation using machine-learning techniques. In this paper we compare different annotation schemes that have been employed by two groups of researchers to annotate large corpora of narrative text. Using a dual-annotation methodology, we investigate the correlation between narrative clauses distinguished by their structural role (orientation, action, evaluation), their subjectivity, and their narrative level within the discourse. We find that each simple narrative annotation scheme captures a structurally distinct characteristic of real-world narratives, and each combination of labels is evident in a corpus of 19 weblog narratives (951 narrative clauses). We discuss several potential applications of minimalist narrative annotation schemes, noting the combination of label across these two annotation schemes that best support each task.

/pdf/minimal-narrative-annotation-schemes-and-their-applications-2ahb85eqta.pdf

Minimal Narrative Annotation Schemes and Their Applications

We address the task of predicting causally related events in stories according to a standard evaluation framework, the Choice of Plausible Alternatives (COPA). We present a neural encoder-decoder model that learns to predict relations between adjacent sequences in stories as a means of modeling causality. We explore this approach using different methods for extracting and representing sequence pairs as well as different model architectures. We also compare the impact of different training datasets on our model. In particular, we demonstrate the usefulness of a corpus not previously applied to COPA, the ROCStories corpus. While not state-of-the-art, our results establish a new reference point for systems evaluated on COPA, and one that is particularly informative for future neural-based approaches.

/pdf/an-encoder-decoder-approach-to-predicting-causal-relations-2marnpk2qn.pdf

An Encoder-decoder Approach to Predicting Causal Relations in Stories

Question-answering dialogue systems have found many applications in interactive learning environments. This paper is concerned with one such application for Army leadership training, where trainees input free-text questions that elicit pre-recorded video responses. Since these responses are already crafted before the question is asked, a certain degree of incoherence exists between the question that is asked and the answer that is given. This paper explores the use of short linking dialogues that stand in between the question and its video response to alleviate the problem of incoherence. We describe a set of experiments with human generated linking dialogues that demonstrate their added value. We then describe our implementation of an automated method for utilizing linking dialogues and show that these have better coherence properties than the original system without linking dialogues.

https://ict.usc.edu/pubs/Improving%20Question-Answering%20With%20Linking%20Dialogues%20.pdf

Andrew S. Gordon

Papers

Strategy Representation: An Analysis of Planning Knowledge

Formalizations of Commonsense Psychology

Minimal Narrative Annotation Schemes and Their Applications

An Encoder-decoder Approach to Predicting Causal Relations in Stories

Improving question-answering with linking dialogues