Philipp Scharpf Profile
Philipp Scharpf

@PhilippScharpf

Followers
24
Following
24
Media
3
Statuses
13

Researcher and Entrepreneur

Germany
Joined April 2021
Don't wanna be here? Send us removal request.
@PhilippScharpf
Philipp Scharpf
2 years
RT @FHKPetersen: I am excited to announce our #ACL2023NLP paper on mathematical formula translation @aclmeeting: .
0
9
0
@PhilippScharpf
Philipp Scharpf
2 years
I am presenting at AIED 2023 - 24th International Conference on Artificial Intelligence in Education. Please check out my talk if you're attending the event @AIEDconf
Tweet media one
0
4
6
@grok
Grok
1 day
Generate videos in just a few seconds. Try Grok Imagine, free for a limited time.
493
256
3K
@PhilippScharpf
Philipp Scharpf
2 years
RT @ana_m_zhukova: We are looking forward talking to you today at #JCDL2023!.
0
4
0
@PhilippScharpf
Philipp Scharpf
2 years
In times of LLMs, such as GPT, verifying the correctness of AI-generated questions is becoming an essential need to ensure quality. PhysWikiQuiz can be employed to check other AI systems. GPT-4 hallucinated values for 30% of the test questions.
Tweet card summary image
researchgate.net
PDF | We present »PhysWikiQuiz«, a physics question generation and test engine. Our system, hosted by Wikimedia, utilizes Wikidata, an open,... | Find, read and cite all the research you need on...
1
2
5
@PhilippScharpf
Philipp Scharpf
2 years
RT @CorneliusIhle: Does your network need an incentive mechanism?. We are excited to announce that our work on peer-to-peer incentive mecha….
0
9
0
@PhilippScharpf
Philipp Scharpf
2 years
RT @MeuschkeN: Excited to present our work on evaluating tools to extract information from academic PDF at #iConf23. Preprint: https://t.co….
0
7
0
@PhilippScharpf
Philipp Scharpf
2 years
The rise of GPT also questions all traditional question answering methods, even in mathematical domains. However, as for natural language, the supply of references still seems to be the unfair advantage of classical MathIR methods:.
Tweet card summary image
arxiv.org
The zbMATH database contains more than 4 million bibliographic entries. We aim to provide easy access to these entries. Therefore, we maintain different index structures, including a formula...
0
3
6
@PhilippScharpf
Philipp Scharpf
3 years
RT @jpwahle: Language models are great. But they are so good at generating text that it can become a severe threat to academic integrity. I….
0
14
0
@PhilippScharpf
Philipp Scharpf
3 years
Checkout our new paper "Collaborative and AI-aided Exam Question Generation using Wikidata" presented @WikidataW #ISWC2022. Paper:.Talk:.System:.
0
2
6
@PhilippScharpf
Philipp Scharpf
3 years
RT @WikiResearch: "Do the Math: Making Mathematics in Wikipedia.Computable" A spell checker for mathematical formulae, tested on Wikipedia….
0
12
0
@PhilippScharpf
Philipp Scharpf
3 years
RT @MeuschkeN: Meinen herzlichsten Dank an alle Beteiligten für diese immense Auszeichnung und die eindrucksvolle Preisverleihung!.
0
1
0
@PhilippScharpf
Philipp Scharpf
3 years
Checkout our new paper on Mathematical Question Answering via Unsupervised Formula Labeling. Talk:.Paper:.
0
0
0