On the Dangers of Stochastic Parrots

Bender, Emily M.; Gebru, Timnit; McMillan-Major, Angelina; Shmitchell, Shmargaret

doi:10.1145/3442188.3445922

Cited by 2,296 publications

(1,036 citation statements)

References 61 publications

Supporting

Mentioning

573

Contrasting

Unclassified

Order By: Relevance

“…This will inevitably follow -and perhaps worsen -familiar disadvantages, both enabling and disenfranchising different groups according to their means. Access will certainly not correlate to need, or environmental impact sustained (Bender et al 2021).…”

Section: Endless Possibilities Vs Boundless Risks Ethical Challengesmentioning

confidence: 99%

The Dawn of the Human-Machine Era: A forecast of new and emerging language technologies

Sayers¹,

Sousa‐Silva²,

Höhn³

et al. 2021

View full text Add to dashboard Cite

New language technologies are coming, thanks to the huge and competing private investment fuelling rapid progress; we can either understand and foresee their effects, or be taken by surprise and spend our time trying to catch up. This report scketches out some transformative new technologies that are likely to fundamentally change our use of language. Some of these may feel unrealistically futuristic or far-fetched, but a central purpose of this report - and the wider LITHME network - is to illustrate that these are mostly just the logical development and maturation of technologies currently in prototype. But will everyone benefit from all these shiny new gadgets? Throughout this report we emphasise a range of groups who will be disadvantaged and issues of inequality. Important issues of security and privacy will accompany new language technologies. A further caution is to re-emphasise the current limitations of AI. Looking ahead, we see many intriguing opportunities and new capabilities, but a range of other uncertainties and inequalities. New devices will enable new ways to talk, to translate, to remember, and to learn. But advances in technology will reproduce existing inequalities among those who cannot afford these devices, among the world’s smaller languages, and especially for sign language. Debates over privacy and security will flare and crackle with every new immersive gadget. We will move together into this curious new world with a mix of excitement and apprehension - reacting, debating, sharing and disagreeing as we always do. Plug in, as the human-machine era dawns.

show abstract

Section: Endless Possibilities Vs Boundless Risks Ethical Challengesmentioning

confidence: 99%

The Dawn of the Human-Machine Era: A forecast of new and emerging language technologies

Sayers¹,

Sousa‐Silva²,

Höhn³

et al. 2021

View full text Add to dashboard Cite

show abstract

“…Indeed, one of our main findings is that there exist unweighted languages ∈ P for which no standard autoregressive model has as its support, i.e., assigns weight > 0 to just the strings x ∈ . This is downright depressing, considering the costs invested in training huge parametric autoregressive models (Bender et al, 2021). Since ∈ P, it is trivial to build an efficient scoring function˜ (x) with fixed parameters that has as its support -just not an autoregressive one.…”

Section: Model Familymentioning

confidence: 99%

Limitations of Autoregressive Models and Their Alternatives

Lin

Jaech

et al. 2021

Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Langua

View full text Add to dashboard Cite

Standard autoregressive language models perform only polynomial-time computation to compute the probability of the next symbol. While this is attractive, it means they cannot model distributions whose next-symbol probability is hard to compute. Indeed, they cannot even model them well enough to solve associated easy decision problems for which an engineer might want to consult a language model. These limitations apply no matter how much computation and data are used to train the model, unless the model is given access to oracle parameters that grow superpolynomially in sequence length.Thus, simply training larger autoregressive language models is not a panacea for NLP. Alternatives include energy-based models (which give up efficient sampling) and latent-variable autoregressive models (which give up efficient scoring of a given string). Both are powerful enough to escape the above limitations.

show abstract

“…In this study, a temporal decline in a specific topic area (e.g., knowledge translation) might indicate that a concept is going out of favor, or simply that the term used to reference the concept has evolved. Analytic approaches involving deep semantic modeling with word vectorization are a step forward though are not without their own limitations [39].…”

Section: Limitationsmentioning

confidence: 99%

5335 days of Implementation Science: using natural language processing to examine publication trends and topics

Scaccia

Scott

2021

Implementation Sci

View full text Add to dashboard Cite

Introduction Moving evidence-based practices into the hands of practitioners requires the synthesis and translation of research literature. However, the growing pace of scientific publications across disciplines makes it increasingly difficult to stay abreast of research literature. Natural language processing (NLP) methods are emerging as a valuable strategy for conducting content analyses of academic literature. We sought to apply NLP to identify publication trends in the journal Implementation Science, including key topic clusters and the distribution of topics over time. A parallel study objective was to demonstrate how NLP can be used in research synthesis. Methods We examined 1711 Implementation Science abstracts published from February 22, 2006, to October 1, 2020. We retrieved the study data using PubMed’s Application Programming Interface (API) to assemble a database. Following standard preprocessing steps, we use topic modeling with Latent Dirichlet allocation (LDA) to cluster the abstracts following a minimization algorithm. Results We examined 30 topics and computed topic model statistics of quality. Analyses revealed that published articles largely reflect (i) characteristics of research, or (ii) domains of practice. Emergent topic clusters encompassed key terms both salient and common to implementation science. HIV and stroke represent the most commonly published clinical areas. Systematic reviews have grown in topic prominence and coherence, whereas articles pertaining to knowledge translation (KT) have dropped in prominence since 2013. Articles on HIV and implementation effectiveness have increased in topic exclusivity over time. Discussion We demonstrated how NLP can be used as a synthesis and translation method to identify trends and topics across a large number of (over 1700) articles. With applicability to a variety of research domains, NLP is a promising approach to accelerate the dissemination and uptake of research literature. For future research in implementation science, we encourage the inclusion of more equity-focused studies to expand the impact of implementation science on disadvantaged communities.

show abstract

On the Dangers of Stochastic Parrots

Cited by 2,296 publications

References 61 publications

The Dawn of the Human-Machine Era: A forecast of new and emerging language technologies

The Dawn of the Human-Machine Era: A forecast of new and emerging language technologies

Limitations of Autoregressive Models and Their Alternatives

5335 days of Implementation Science: using natural language processing to examine publication trends and topics

Contact Info

Product

Resources

About