article thumbnail

What are they thinking?

Allen AI

2020) and Macaw (Tafjord and Clark, 2021), our results show that mental models derived using these LMs’ predictions are significantly inconsistent, with 19–43% conditional violation. Association for Computational Linguistics. Association for Computational Linguistics. Association for Computational Linguistics.

article thumbnail

AI2 at ACL 2023

Allen AI

HINT models outperform strong state-of-the-art baselines by over 10% when controlling for compute (measured in FLOPs). Submissions to EMNLP 2021 binned by count of YES responses to the NLP Reproducibility Checklist items. 2021) and summarization (Saunders et al., The ACCEPT rate is given for each bin.

professionals

Sign Up for our Newsletter

This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.

article thumbnail

SQuARE: Towards Multi-Domain and Few-Shot Collaborating Question Answering Agents

ODSC - Open Data Science

Euro) in 2021. She is currently the president of the Association of Computational Linguistics. Examples are the ACL fellow award 2020 and the first Hessian LOEWE Distinguished Chair award (2,5 mil. Iryna is co-director of the NLP program within ELLIS, a European network of excellence in machine learning.

article thumbnail

ClarifyDelphi

Allen AI

2021), a recently proposed commonsense moral reasoning model, generates moral judgments for simple actions described in text. arXiv preprint arXiv:2110.07574 (2021). In Findings of the Association for Computational Linguistics: EMNLP 2020 , pp. If offered to a work colleague, it may even be viewed as a courteous gesture.

article thumbnail

ML and NLP Research Highlights of 2021

Sebastian Ruder

2021) 2021 saw many exciting advances in machine learning (ML) and natural language processing (NLP).   2021 saw the continuation of the development of ever larger pre-trained models. 2021 saw the development of alternative model architectures that are viable alternatives to the transformer. style loss.

NLP 52
article thumbnail

The State of Multilingual AI

Sebastian Ruder

For a recent study [3] , we similarly reviewed papers from ACL 2021 and found that almost 70% of papers only evaluate on English. Initiatives   The Association for Computational Linguistics (ACL) has emphasized the importance of language diversity, with a special theme track at the main ACL 2022 conference on this topic.

article thumbnail

Selective Classification Can Magnify Disparities Across Groups

The Stanford AI Lab Blog

This post is based off our ICLR 2021 paper : Selective Classification Can Magnify Disparities Across Groups. In Association for Computational Linguistics (ACL), pp. arXiv preprint arXiv:2102.10395, 2021. ↩ Erik Jones*, Shiori Sagawa* Pang Wei Koh*, Ananya Kumar, and Percy Liang. In World Wide Web (WWW), pp.