Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data

Bender, Emily M.; Koller, Alexander

doi:10.18653/v1/2020.acl-main.463

Cited by 571 publications

(480 citation statements)

References 44 publications

Supporting

Mentioning

366

Contrasting

Unclassified

Order By: Relevance

“…Interestingly, GPT-3 also performs quite well on semantic and discourse reasoning tasks, such as resolving pronouns or predicting final words of sentences. In light of the impressive achievements of GPT-3, an interesting discussion has been unfolding regarding the ability of such language models to present a “general AI,” and whether the often attested lack of meaning or understanding of such models that mainly operate at the level of forms ( Bender and Koller, 2020 ) is even relevant. 1 Indeed, the fascinating performance in many tasks demonstrates that many patterns underlying semantically coherent language use can be extracted and synthesized when scaling up the models and the data.…”

Section: Computational Models Of Human-agent Communicationmentioning

confidence: 99%

Revisiting Human-Agent Communication: The Importance of Joint Co-construction and Understanding Mental States

Kopp

Krämer

2021

Front. Psychol.

View full text Add to dashboard Cite

The study of human-human communication and the development of computational models for human-agent communication have diverged significantly throughout the last decade. Yet, despite frequently made claims of “super-human performance” in, e.g., speech recognition or image processing, so far, no system is able to lead a half-decent coherent conversation with a human. In this paper, we argue that we must start to re-consider the hallmarks of cooperative communication and the core capabilities that we have developed for it, and which conversational agents need to be equipped with: incremental joint co-construction and mentalizing. We base our argument on a vast body of work on human-human communication and its psychological processes that we reason to be relevant and necessary to take into account when modeling human-agent communication. We contrast those with current conceptualizations of human-agent interaction and formulate suggestions for the development of future systems.

show abstract

Section: Computational Models Of Human-agent Communicationmentioning

confidence: 99%

Revisiting Human-Agent Communication: The Importance of Joint Co-construction and Understanding Mental States

Kopp

Krämer

2021

Front. Psychol.

View full text Add to dashboard Cite

show abstract

“…Of course, there is much to dislike about BERT and its ilk, primarily the fact that it lacks not only any kind of communicative goals, but any links to real-world meanings at all (Bender & Koller, 2020): Words are represented as vectors that capture their distributional similarity to other words (a kind of souped-up Latent Semantic Analysis), albeit in a context-dependent fashion (e.g. table would have different vectors in the input string He sat at the table and See Table 1 for details ).…”

Section: Can You Tell Me How To Get How To Get To Abstract Reprmentioning

confidence: 99%

Abstractions made of exemplars or ‘You’re all right, and I’ve changed my mind’: Response to commentators

Ambridge

2020

First Language

View full text Add to dashboard Cite

In this response to commentators, I agree with those who suggested that the distinction between exemplar- and abstraction-based accounts is something of a false dichotomy and therefore move to an abstractions-made-of-exemplars account under which (a) we store all the exemplars that we hear (subject to attention, decay, interference, etc.) but (b) in the service of language use, re-represent these exemplars at multiple levels of abstraction, as simulated by computational neural-network models such as BERT, ELMo and GPT-3. Whilst I maintain that traditional linguistic abstractions (e.g. a DETERMINER category; SUBJECT VERB OBJECT word order) are no more than human-readable approximations of the type of abstractions formed by both human and artificial multiple-layer networks, I express hope that the abstractions-made-of-exemplars position can point the way towards a truce in the language acquisition wars: We were all right all along, just focusing on different levels of abstraction.

show abstract

“…Moreover, Ettinger (2020) found that the popular BERT model (Devlin et al, 2019) completely failed to acquire a general understanding of negation. Related, Bender and Koller (2020) contend that meaning cannot be learned from form alone, and argue for approaches that focus on grounding the language (communication) in the real world. We believe formal meaning representations therefore have an important role to play in future semantic applications, as semantic parsers produce an explicit model of a real-world interpretation.…”

Section: Introductionmentioning

confidence: 99%

Character-level Representations Improve DRS-based Semantic Parsing Even in the Age of BERT

Noord¹,

Toral²,

Bos³

2020

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)

View full text Add to dashboard Cite

We combine character-level and contextual language model representations to improve performance on Discourse Representation Structure parsing. Character representations can easily be added in a sequence-to-sequence model in either one encoder or as a fully separate encoder, with improvements that are robust to different language models, languages and data sets. For English, these improvements are larger than adding individual sources of linguistic information or adding non-contextual embeddings. A new method of analysis based on semantic tags demonstrates that the character-level representations improve performance across a subset of selected semantic phenomena.

show abstract

Climbing towards NLU: On Meaning, Form, and Understanding in the Age of Data

Cited by 571 publications

References 44 publications

Revisiting Human-Agent Communication: The Importance of Joint Co-construction and Understanding Mental States

Revisiting Human-Agent Communication: The Importance of Joint Co-construction and Understanding Mental States

Abstractions made of exemplars or ‘You’re all right, and I’ve changed my mind’: Response to commentators

Character-level Representations Improve DRS-based Semantic Parsing Even in the Age of BERT

Contact Info

Product

Resources

About