Niklas Stoehr
@niklasstoehr.bsky.social
📤 1053
📥 213
📝 5
Gemini Post-Training ⚫️ Research Scientist at Google DeepMind ⚫️ PhD from ETH Zurich
reposted by
Niklas Stoehr
Alexander Hoyle
19 days ago
Paper:
arxiv.org/abs/2509.03116
Code:
github.com/haukelicht/s..
. With:
@haukelicht.bsky.social
*
@rupak-s.bsky.social
*
@patrickwu.bsky.social
@pranavgoel.bsky.social
@niklasstoehr.bsky.social
@elliottash.bsky.social
loading . . .
https://github.com/haukelicht/s..
0
3
1
reposted by
Niklas Stoehr
Alexander Hoyle
19 days ago
[corrected link] LLMs are often used for text annotation in social science. In some cases, this involves placing text items on a scale: eg, 1 for liberal and 9 for conservative There are a few ways to handle this task. Which work best? Our new EMNLP paper has some answers🧵
arxiv.org/abs/2509.03116
1
24
5
reposted by
Niklas Stoehr
Alexander Hoyle
4 months ago
Evaluating topic models (and document clustering methods) is hard. In fact, since our paper critiquing standard evaluation practices four years ago, there hasn't been a good replacement metric That ends today (we hope)! Our new ACL paper introduces an LLM-based evaluation protocol 🧵
3
52
12
🎓 I recently defended my PhD and moved from one dream team at ETH Zurich to another at DeepMind—a huge thank you to the many people who have supported me along the way!
5 months ago
0
31
0
reposted by
Niklas Stoehr
Shauli Ravfogel
9 months ago
Our paper "A Practical Method for Generating String Counterfactuals" has been accepted to the findings of NAACL 2025! a joint work with
@matan-avitan.bsky.social
,
@yoavgo.bsky.social
and Ryan Cotterell. We propose "Intervention Lens", a technique to explain intervention in natural language. (1/6)
1
38
6
reposted by
Niklas Stoehr
Paul Röttger @ EMNLP
9 months ago
Are LLMs biased when they write about political issues? We just released IssueBench – the largest, most realistic benchmark of its kind – to answer this question more robustly than ever before. Long 🧵with spicy results 👇
4
84
31
reposted by
Niklas Stoehr
Julian Minder
12 months ago
Can we understand and control how language models balance context and prior knowledge? Our latest paper shows it’s all about a 1D knob! 🎛️
arxiv.org/abs/2411.07404
Co-led with
@kevdududu.bsky.social
-
@niklasstoehr.bsky.social
, Giovanni Monea,
@wendlerc.bsky.social
, Robert West & Ryan Cotterell.
1
13
3
reposted by
Niklas Stoehr
Lucy Li
12 months ago
mech interp:
bsky.app/starter-pack...
women in nlp:
bsky.app/starter-pack...
nlp #1:
bsky.app/starter-pack...
nlp #2:
bsky.app/starter-pack...
ml/data/tech:
bsky.app/starter-pack...
robotics & ai:
bsky.app/starter-pack...
add a skeleton here at some point
7
74
23
reposted by
Niklas Stoehr
Sweta Karlekar
12 months ago
If you’re interested in mechanistic interpretability, I just found this starter pack and wanted to boost it (thanks for creating it
@butanium.bsky.social
!). Excited to have a mech interp community on bluesky 🎉
go.bsky.app/LisK3CP
add a skeleton here at some point
3
36
10
reposted by
Niklas Stoehr
Giuliano Formisano
12 months ago
Just launched a Political Comm/NLP/Text-as-Data Starter Pack. 🦋🤗 Join us and/or drop a message to be added!
go.bsky.app/39MWTjg
#starterpack
#polsci
add a skeleton here at some point
3
30
10
reposted by
Niklas Stoehr
Vilém Zouhar #EMNLP
12 months ago
Trying to bring ML/NLP/etal people from ETH Zürich together. Ping me to add you. 🙂
bsky.app/starter-pack...
add a skeleton here at some point
1
26
6
you reached the end!!
feeds!
log in