Tim Baumgรคrtner
@timbmg.bsky.social
๐ค 85
๐ฅ 263
๐ 3
๐จโ๐ป NLP PhD Student
@ukplab.bsky.social
๐ก TIL
@overleaf.com
is basically a git repo. In my research workflow, I directly added it as submodule to my code repo. Now I can produce figures and tables, and have them magically uploaded to Overleaf just by pushing the repo. No more renaming, keeping versions straight, and manual uploading ๐
about 1 month ago
0
0
0
๐ก TIL, it's super easy to fetch data from Google Sheets into Pandas. Makes it really convenient to annotate some data. Previously, I was always downloading CSVs, losing track of file versions, and loading and merging them sluggishly in Python. ๐ find the code here:
gist.github.com/timbmg/6c2d6...
about 2 months ago
0
0
0
reposted by
Tim Baumgรคrtner
UKP Lab
8 months ago
๐ ๐ช๐ฎ๐ป๐ ๐๐ผ ๐ฒ๐๐ฎ๐น๐๐ฎ๐๐ฒ ๐บ๐ผ๐ฑ๐ฒ๐น๐ ๐ผ๐ป ๐๐ฐ๐ถ๐ฒ๐ป๐๐ถ๐ณ๐ถ๐ฐ ๐ค๐, ๐ฏ๐๐ ๐๐ผ๐๐ฟ ๐ฑ๐ฎ๐๐ฎ๐๐ฒ๐ ๐น๐ฎ๐ฐ๐ธ๐ ๐ฟ๐ฒ๐ฎ๐น-๐๐ผ๐ฟ๐น๐ฑ ๐พ๐๐ฒ๐๐๐ถ๐ผ๐ป๐ ๐ฎ๐๐ธ๐ฒ๐ฑ ๐ฏ๐ ๐ฒ๐ ๐ฝ๐ฒ๐ฟ๐๐? ๐ PeerQA is the solution: a dataset with questions from peer reviews and answers from the original authors. (1/๐งต)
#NLProc
1
2
1
reposted by
Tim Baumgรคrtner
Martin Tutek
10 months ago
๐จ๐จ New preprint ๐จ๐จ Ever wonder whether verbalized CoTs correspond to the internal reasoning process of the model? We propose a novel parametric faithfulness approach, which erases information contained in CoT steps from the model parameters to assess CoT faithfulness.
arxiv.org/abs/2502.14829
loading . . .
Measuring Faithfulness of Chains of Thought by Unlearning Reasoning Steps
When prompted to think step-by-step, language models (LMs) produce a chain of thought (CoT), a sequence of reasoning steps that the model supposedly used to produce its prediction. However, despite mu...
https://arxiv.org/abs/2502.14829
2
47
16
reposted by
Tim Baumgรคrtner
UKP Lab
10 months ago
๐๐ฎ๐ฐ๐-๐๐ต๐ฒ๐ฐ๐ธ๐ถ๐ป๐ด ๐ถ๐ป ๐๐ต๐ฒ ๐๐ด๐ฒ ๐ผ๐ณ ๐๐ โ ๐ ๐ง๐ฎ๐น๐ธ ๐ฏ๐ ๐๐ฟ๐๐ป๐ฎ ๐๐๐ฟ๐ฒ๐๐๐ฐ๐ต @๐๐ ๐ณ๐ผ๐ฟ ๐๐ผ๐ผ๐ฑ Misinformation is a new weapon disrupting public debates, scientific discussions, and political decisions. How can we identify and counter misleading content? (1/๐งต)
loading . . .
Towards real-world fact-checking with large language models
Misinformation poses a growing threat to our society. It has a severe impact on public health by promoting fake cures fear and distrust. Current research
https://aiforgood.itu.int/event/towards-real-world-fact-checking-with-large-language-models/
1
3
1
reposted by
Tim Baumgรคrtner
Florent Daudens
11 months ago
๐ค An Energy Star for AI? Introducing AI Energy Score: First-ever rating system comparing 166 AI models' energy consumption! From LLaMa to Gemma, get transparent โญ๏ธ1-5 efficiency ratings. Incredible work led by
@sashamtl.bsky.social
huggingface.co/blog/sasha/a...
0
25
8
Excited to share that our Paper "PeerQA: A Scientific Question Answering Dataset from Peer Reviews" as been accepted to
#NAACL2025
Looking forward to presenting it in Albuquerque ๐๏ธ!
add a skeleton here at some point
11 months ago
0
5
0
reposted by
Tim Baumgรคrtner
Bertram Hรธjer
about 1 year ago
What do YOU mean by "intelligence", and does ChatGPT fit your definition? We collected the major criteria used in CogSci and other fields, and designed a survey to find out! Access link:
www.survey-xact.dk/collect
Code: 4S7V-SN4M-S536 Time: 5-10 mins
loading . . .
Perspectives on Intelligence: Community Survey
Research survey exploring how NLP/ML/CogSci researchers define and use the concept of intelligence.
https://bertramhojer.github.io/projects/intelligence-survey/
2
32
23
you reached the end!!
feeds!
log in