It's here!
Upload *any paper* to Explainpaper and start instantly getting explanations! Ask follow up questions if you need a more in-depth answer.
Go to explainpaper.com and go read all the papers you've been saving! πππ
Conversation
Replying to
Would love to see something like this, but a scoring system that evaluates the quality of the study behind the paper (sufficient N, sufficient participant diversity, double-blind RCT, replicability, etc.)
1
2
3
Show replies
Replying to
Thank you for using ours as an example :) Hope to share our follow-up work soon: βlarge language model (papers) can self-explainβ
Quote Tweet
(1/8) *new paper* βLLMs can self-improveβ
w/ *self-generated CoTs* (βlogical dark knowledgeβ), no GT labels:
- SoTA (74.4%->82.1% GSM8K, 90.0%->94.4% OpenBookQA, 63.4%->67.9% ANLI-A3) by fine-tuning
- SoTA βzero-shotβ (GSM8K 70.1% -> 74.2%) by prompting
arxiv.org/abs/2210.11610
Show this thread
1
2
14
Replying to
Quick issue to be aware of. If you select a large string of text, the followup question box disappears (unless you zoom out a crazy amount).
1
1
Replying to
yes, this is an issue. we did not originally design this with the notion that people would select large amounts of text (but they do!)
1
Replying to
As a test, have you found any authors willing to give it a try and compare their mental maps to the output?
2
1
16
Replying to
Yeah! Interestingly enough a lot of authors start out by uploading their one papers π
1
2
10
Show replies