Jay DeYoung

YI @ AI2

I am sure that my employer does not endorse anything I say.

[Lucy Lu Wang @lucyluwang]

Sharing our #ACL2023NLP paper on evaluation for medical multi-document summarization! New human annotated dataset, new metrics, and an in-depth analysis, here: https://t.co/QpyE6B1WoH

Joint w/ @YuliaOtmakhova @jaydepun @ththinh_ BaileyKuehl ErinBransom @allen_ai @byron_c_wallace

[Jay DeYoung @jaydepun]

@NirRatner @AI21Labs Something similar works for multidocument summarization: I trained an encoder/decoder model to independently encode inputs/concat/decode:
https://t.co/Bv5ONOIMi5

[Jay DeYoung @jaydepun]

@jayvanbavel This is not a realistic proposition. They've all lost weight.

[Jay DeYoung @jaydepun]

AI safety will be an important part of any system performing these tasks in the wild. There’s a lot of work to do to ensure the quality and reliability of model outputs. We encourage the community to work on these challenging and important problems! 3/3

[Jay DeYoung @jaydepun]

Medical systematic reviews are costly and time-consuming to produce. We introduce a new dataset called MS^2 to help automate and assist in parts of the process: https://t.co/NKFJsklCJY #NLProc @lucyluwang @i_beltagy @SemanticScholar @allen_ai 1/3

[Jay DeYoung @jaydepun]

MS^2 focuses on extraction and summarization in the review pipeline. We harvest 20K systematic reviews and 470K of their references from Semantic Scholar, identify summary targets, and experiment with multi-document summarization methods. 2/3

[Jan-Willem van de Meent @jwvdm]

1/ New work by Alican (@alicanb_) and Babak (@BabakEsmaeili10): "Evaluating Combinatorial Generalization in Variational Autoencoders" (https://t.co/XtB29f9FVV)

In this paper we ask the question: "To what extent do VAEs generalize to unseen combinations of features?"(thread)

[Hao Wu @Hao_Wu_]

Happy to share our work on Amortized Population Gibbs Samplers! arXiv: https://t.co/lxchFpVUA1

[Nazneen Rajani @nazneenrajani]

#NLProc does not have a standard benchmark for interpretability. I am stoked to announce ERASER: the first-ever effort on unifying and standardizing NLP tasks with the goal of interpretability.
https://t.co/8l74WNiYzm

[Jay DeYoung @jaydepun]

We decided to try to measure it: https://t.co/w12i09aac3

@byron_c_wallace @successar_nlp @nazneenrajani @RichardSocher

#NLProc #MachineLearning

[Jay DeYoung @jaydepun]

How Rationale are your ML Models?
#NLProc #MachineLearning

[Jay DeYoung @jaydepun]

We have no idea.