English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Tools annual

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Many real-world problems require the combined application of multiplereasoning abilities employing suitable abstractions, commonsense knowledge, andcreative synthesis of problem-solving strategies. To help advance AI systemstowards such capabilities, we propose a new reasoning challenge, namely FermiProblems (FPs), which are questions whose answers can only be approximatelyestimated because their precise computation is either impractical orimpossible. For example, "How much would the sea level rise if all ice in theworld melted?" FPs are commonly used in quizzes and interviews to bring out andevaluate the creative reasoning abilities of humans. To do the same for AIsystems, we present two datasets: 1) A collection of 1k real-world FPs sourcedfrom quizzes and olympiads; and 2) a bank of 10k synthetic FPs of intermediatecomplexity to serve as a sandbox for the harder real-world challenge. Inaddition to question answer pairs, the datasets contain detailed solutions inthe form of an executable program and supporting facts, helping in supervisionand evaluation of intermediate steps. We demonstrate that even extensivelyfine-tuned large scale language models perform poorly on these datasets, onaverage making estimates that are off by two orders of magnitude. Ourcontribution is thus the crystallization of several unsolved AI problems into asingle, new challenge that we hope will spur further advances in buildingsystems that can reason.

How Much Coffee Was Consumed During EMNLP 2019? Fermi Problems: A New Reasoning Challenge for AI