Daniel Braun
@d4br4.bsky.social
Professor of Computer Science at the University of Marburg. Working on Natural Language Processing (NLP), AI for Social Good, and Legal Tech.
https://www.responsible-nlp.net
https://www.responsible-nlp.net
Alles Gute zum #GIburtstag! Als GI-Mitglieder gestalten wir gemeinsam die digitale Zukunft. 💛
Lust mitzumachen? Hier gibt's alle Infos zur Mitgliedschaft: gi.de/mitgliedschaft #WirSindInformatik
Lust mitzumachen? Hier gibt's alle Infos zur Mitgliedschaft: gi.de/mitgliedschaft #WirSindInformatik
September 16, 2025 at 8:01 AM
Alles Gute zum #GIburtstag! Als GI-Mitglieder gestalten wir gemeinsam die digitale Zukunft. 💛
Lust mitzumachen? Hier gibt's alle Infos zur Mitgliedschaft: gi.de/mitgliedschaft #WirSindInformatik
Lust mitzumachen? Hier gibt's alle Infos zur Mitgliedschaft: gi.de/mitgliedschaft #WirSindInformatik
This week, I’m in Heidelberg attending the Heidelberg Laureat Forum #hlf25. It’s always a great pleasure to be among some of the smartest minds in Math and Computer Science and see what they are working on.
September 15, 2025 at 8:55 AM
This week, I’m in Heidelberg attending the Heidelberg Laureat Forum #hlf25. It’s always a great pleasure to be among some of the smartest minds in Math and Computer Science and see what they are working on.
Are LLMs “Yes Sayers”?
Humans often show acquiescence bias, i.e. prefer positive answers in questionnaires and surveys regardless of their actual belief. We investigated if LLMs show a similar bias and found: they’re disproportionately likely to answer “No,” even when the meaning flips.
Humans often show acquiescence bias, i.e. prefer positive answers in questionnaires and surveys regardless of their actual belief. We investigated if LLMs show a similar bias and found: they’re disproportionately likely to answer “No,” even when the meaning flips.
Acquiescence Bias in Large Language Models
Acquiescence bias, i.e. the tendency of humans to agree with statements in surveys, independent of their actual beliefs, is well researched and documented. Since Large Language Models (LLMs) have…
arxiv.org
September 12, 2025 at 10:39 AM
Are LLMs “Yes Sayers”?
Humans often show acquiescence bias, i.e. prefer positive answers in questionnaires and surveys regardless of their actual belief. We investigated if LLMs show a similar bias and found: they’re disproportionately likely to answer “No,” even when the meaning flips.
Humans often show acquiescence bias, i.e. prefer positive answers in questionnaires and surveys regardless of their actual belief. We investigated if LLMs show a similar bias and found: they’re disproportionately likely to answer “No,” even when the meaning flips.
Reposted by Daniel Braun
Excited to announce the first-ever Workshop for Young Researchers in Natural Language Generation (YNLG), supported by @siggen.bsky.social, taking place on October 29, 2025 in Hanoi, Vietnam, co-located with INLG 2025.
Call for Submissions is out now!
ynlg-workshop.github.io
Call for Submissions is out now!
ynlg-workshop.github.io
August 12, 2025 at 7:05 AM
Excited to announce the first-ever Workshop for Young Researchers in Natural Language Generation (YNLG), supported by @siggen.bsky.social, taking place on October 29, 2025 in Hanoi, Vietnam, co-located with INLG 2025.
Call for Submissions is out now!
ynlg-workshop.github.io
Call for Submissions is out now!
ynlg-workshop.github.io
On my way to Vienna for #ACL2025 @aclmeeting.bsky.social. Hit me up if you want to talk about Legal NLP, Perspectivism and Disagreement, AI for Social Good or other stuff.
July 26, 2025 at 1:22 PM
On my way to Vienna for #ACL2025 @aclmeeting.bsky.social. Hit me up if you want to talk about Legal NLP, Perspectivism and Disagreement, AI for Social Good or other stuff.
Reposted by Daniel Braun
Ive written a "Last Word" opinion piece for CL about evaluating real-world impact. It
* looks at how impact can be evaluated
* shows via a structured survey that perhaps 0.1% of ACL Anth papers measure real-world impact
* discusses why this is the case
arxiv.org/abs/2507.05973
* looks at how impact can be evaluated
* shows via a structured survey that perhaps 0.1% of ACL Anth papers measure real-world impact
* discusses why this is the case
arxiv.org/abs/2507.05973
We Should Evaluate Real-World Impact
The ACL community has very little interest in evaluating the real-world impact of NLP systems. A structured survey of the ACL Anthology shows that perhaps 0.1% of its papers contain such evaluations; ...
arxiv.org
July 9, 2025 at 8:15 AM
Ive written a "Last Word" opinion piece for CL about evaluating real-world impact. It
* looks at how impact can be evaluated
* shows via a structured survey that perhaps 0.1% of ACL Anth papers measure real-world impact
* discusses why this is the case
arxiv.org/abs/2507.05973
* looks at how impact can be evaluated
* shows via a structured survey that perhaps 0.1% of ACL Anth papers measure real-world impact
* discusses why this is the case
arxiv.org/abs/2507.05973
Now finally assigned an issue number: my article »Why “Artificial Intelligence” Should Not Be Regulated«, in which I argue not against technology regulation, but against arbitrary definitions of 'AI' by lawmakers and in favor of risk-based, technology-independent regulation.
Why “Artificial Intelligence” Should Not Be Regulated | Digital Government: Research and Practice
Lawmakers all over the world have started to draft new regulations for Artificial
Intelligence (AI). While the European Union is currently leading the way with its
AI Act, many other legislators will ...
dl.acm.org
June 17, 2025 at 9:13 AM
Now finally assigned an issue number: my article »Why “Artificial Intelligence” Should Not Be Regulated«, in which I argue not against technology regulation, but against arbitrary definitions of 'AI' by lawmakers and in favor of risk-based, technology-independent regulation.
Reposted by Daniel Braun
Hosted at RUB, with people attending from Frankfurt, Darmstadt, Regensburg, Magdeburg, and Prague, I think our AI & Law Seminars is growing pretty well! Thanks @d4br4.bsky.social for coming over and giving a super inspiring talk on "disagreement (?)" in legal NLP
May 7, 2025 at 3:37 PM
Hosted at RUB, with people attending from Frankfurt, Darmstadt, Regensburg, Magdeburg, and Prague, I think our AI & Law Seminars is growing pretty well! Thanks @d4br4.bsky.social for coming over and giving a super inspiring talk on "disagreement (?)" in legal NLP
Spent a very interesting day talking about NLP in the legal domain at @ruhr-uni-bochum.de. Thank you for having me and the nice discussions!
Looking forward to our next AI & Law Seminar later today with Daniel Braun. Interested in joining vitually? Here's the zoom link: ruhr-uni-bochum.zoom-x.de/j/6645491155...
Meeting ID: 664 5491 1552
Passcode: 462718
Meeting ID: 664 5491 1552
Passcode: 462718
May 7, 2025 at 1:41 PM
Spent a very interesting day talking about NLP in the legal domain at @ruhr-uni-bochum.de. Thank you for having me and the nice discussions!
Reposted by Daniel Braun
🏹 Job alert: DTU Compute is hiring a Tenure Track Assistant/Associate Professor in AI, funded by the Pioneer Centre for AI (@aicentre.dk). DTU Compute is a key partner in the Unit Copenhagen.
📍 Kgs. Lyngby, near Copenhagen
📅 Apply by Apr 30
🔗 More info: efzu.fa.em2.oraclecloud.com/hcmUI/Candid...
📍 Kgs. Lyngby, near Copenhagen
📅 Apply by Apr 30
🔗 More info: efzu.fa.em2.oraclecloud.com/hcmUI/Candid...
Associate Professor or DTU Tenure Track Assistant Professor in AI - DTU Compute
Are you passionate about Human Centered AI and can you help build the next generations of AI systems within the Danish Pioneer Centre for AI? DTU Compute has an attractive group leader opening as Tenu...
efzu.fa.em2.oraclecloud.com
March 31, 2025 at 2:28 PM
🏹 Job alert: DTU Compute is hiring a Tenure Track Assistant/Associate Professor in AI, funded by the Pioneer Centre for AI (@aicentre.dk). DTU Compute is a key partner in the Unit Copenhagen.
📍 Kgs. Lyngby, near Copenhagen
📅 Apply by Apr 30
🔗 More info: efzu.fa.em2.oraclecloud.com/hcmUI/Candid...
📍 Kgs. Lyngby, near Copenhagen
📅 Apply by Apr 30
🔗 More info: efzu.fa.em2.oraclecloud.com/hcmUI/Candid...
Verwaltungsanalogisierung
March 25, 2025 at 11:46 AM
Verwaltungsanalogisierung
Reposted by Daniel Braun
We are hiring for PhD positions! I’m looking for people interested in exploring the intersection of learning and reasoning with applications to anomaly detection and sports. @dtai-kuleuven.bsky.social @wannesm.bsky.social
www.kuleuven.be/personeel/jo...
www.kuleuven.be/personeel/jo...
Several PhD Positions in Artificial Intelligence at Computer Science KU Leuven
All positions are in the Machine Learning subgroup of the Section for Declarative Languages and Artificial Intelligence (DTAI), which is part of the Department of Computer Science at KU Leuven. The DT...
www.kuleuven.be
February 11, 2025 at 7:39 PM
We are hiring for PhD positions! I’m looking for people interested in exploring the intersection of learning and reasoning with applications to anomaly detection and sports. @dtai-kuleuven.bsky.social @wannesm.bsky.social
www.kuleuven.be/personeel/jo...
www.kuleuven.be/personeel/jo...
Reposted by Daniel Braun
Come present your #NLP work in Leuven!
🎈 The CLIN35 Call for Abstracts is out!
clin35.ccl.kuleuven.be/call-for-abs...
The conference on Computational Linguistics in the Netherlands will take place in the ❤️ of Leuven. We're happy to host you on September 12th! 🇧🇪
Looking forward to read and hear more about your recent work at #CLIN35!
clin35.ccl.kuleuven.be/call-for-abs...
The conference on Computational Linguistics in the Netherlands will take place in the ❤️ of Leuven. We're happy to host you on September 12th! 🇧🇪
Looking forward to read and hear more about your recent work at #CLIN35!
CLIN35 - Call for Abstracts
We invite submissions for CLIN35, the 35th edition of the Computational Linguistics in the Netherlands (CLIN) conference, which will take place in Leuven on September 12th, 2025.
Abstracts describing ...
clin35.ccl.kuleuven.be
February 25, 2025 at 7:49 AM
Come present your #NLP work in Leuven!
A hill I will die on is that AI ⊃ ML and not AI = ML. If regulators only mean to regulate ML they should simply do so and not try to redefine AI.
February 17, 2025 at 6:32 PM
A hill I will die on is that AI ⊃ ML and not AI = ML. If regulators only mean to regulate ML they should simply do so and not try to redefine AI.
Reposted by Daniel Braun
🏰 PhD position on computational modelling | collective dynamics | cultural evolution 🏰
Get in touch with any questions you might have and join me in beautiful Marburg!
stellenangebote.uni-marburg.de/jobposting/c... (Deutsch)
stellenangebote.uni-marburg.de/jobposting/3... (English)
Get in touch with any questions you might have and join me in beautiful Marburg!
stellenangebote.uni-marburg.de/jobposting/c... (Deutsch)
stellenangebote.uni-marburg.de/jobposting/3... (English)
February 3, 2025 at 8:52 AM
🏰 PhD position on computational modelling | collective dynamics | cultural evolution 🏰
Get in touch with any questions you might have and join me in beautiful Marburg!
stellenangebote.uni-marburg.de/jobposting/c... (Deutsch)
stellenangebote.uni-marburg.de/jobposting/3... (English)
Get in touch with any questions you might have and join me in beautiful Marburg!
stellenangebote.uni-marburg.de/jobposting/c... (Deutsch)
stellenangebote.uni-marburg.de/jobposting/3... (English)
Reposted by Daniel Braun
Over the last few years, several people from the @milanlp.bsky.social lab went onto the academic job market in #NLP. Based on the conversations with them, I collected some observations we shared internally. In case they are useful for others, though, here they are:
dirkhovy.com/post/2025_01...
dirkhovy.com/post/2025_01...
How to professor | Dirk Hovy
It feels not long ago that I was worried about my job applications, but in reality, I have done this long enough that some of my first MSc supervisees are professors of their own now. Every few months...
dirkhovy.com
January 24, 2025 at 4:25 PM
Over the last few years, several people from the @milanlp.bsky.social lab went onto the academic job market in #NLP. Based on the conversations with them, I collected some observations we shared internally. In case they are useful for others, though, here they are:
dirkhovy.com/post/2025_01...
dirkhovy.com/post/2025_01...
I always find the notion that the European AI industry is held back by (over) regulation quite surprising. Most rules of the AI act will not apply before August 2026. GDPR came into force in 2018. Where have all the EU's big tech companies been before? Clearly regulation is not our (biggest) problem
January 23, 2025 at 8:59 PM
I always find the notion that the European AI industry is held back by (over) regulation quite surprising. Most rules of the AI act will not apply before August 2026. GDPR came into force in 2018. Where have all the EU's big tech companies been before? Clearly regulation is not our (biggest) problem
I'm hiring! Fully funded, full-time PhD position in Natural Language Processing at the University of Marburg. It's an open topic position, so you will have the chance to shape your own topic within the research areas of the group. If you have any questions or doubts please get in touch. #NLP #NLProc
Doctoral Researcher/PhD
stellenangebote.uni-marburg.de
January 6, 2025 at 12:41 PM
I'm hiring! Fully funded, full-time PhD position in Natural Language Processing at the University of Marburg. It's an open topic position, so you will have the chance to shape your own topic within the research areas of the group. If you have any questions or doubts please get in touch. #NLP #NLProc
Very happy to see our paper on the ABG-DE corpus for the legal assessment of clauses in German contracts on the list!
aclanthology.org/2024.acl-lon...
aclanthology.org/2024.acl-lon...
New blog: Interesting papers in 2024
Lots of interesting papers in 2024. I describe a few, and also list others I have mentioned in previous blogs. All are about evaluation, experimental rigour, real-world utility, and/or healthcare applications.
ehudreiter.com/2024/12/16/i...
Lots of interesting papers in 2024. I describe a few, and also list others I have mentioned in previous blogs. All are about evaluation, experimental rigour, real-world utility, and/or healthcare applications.
ehudreiter.com/2024/12/16/i...
Interesting papers in 2024
There were lots of interesting papers in 2024. I describe a few of them, and also list others I have mentioned in previous blogs; all are about evaluation, experimental rigour, real-world utility, …
ehudreiter.com
December 16, 2024 at 8:20 AM
Very happy to see our paper on the ABG-DE corpus for the legal assessment of clauses in German contracts on the list!
aclanthology.org/2024.acl-lon...
aclanthology.org/2024.acl-lon...
Exciting career news: Next month, I will join the University of Marburg as Professor for Computer Science and Head of the Natural Language Processing group. That also means I will be hiring PhD students soon, details will follow.
November 1, 2024 at 2:05 PM
Exciting career news: Next month, I will join the University of Marburg as Professor for Computer Science and Head of the Natural Language Processing group. That also means I will be hiring PhD students soon, details will follow.
Why "Artificial Intelligence" Should Not Be Regulated
For a special issue of Digital Government, I wrote a commentary arguing that "AI" should not be regulated. Not because AI can't be dangerous or AI systems shouldn't be regulated, but because AI is ill-defined and a useless regulatory category.
For a special issue of Digital Government, I wrote a commentary arguing that "AI" should not be regulated. Not because AI can't be dangerous or AI systems shouldn't be regulated, but because AI is ill-defined and a useless regulatory category.
Why
Lawmakers all over the world have started to draft new regulations for Artificial
Intelligence (AI). While the European Union is currently leading the way with its
AI Act, many other legislators will ...
dl.acm.org
September 16, 2024 at 1:12 PM
Why "Artificial Intelligence" Should Not Be Regulated
For a special issue of Digital Government, I wrote a commentary arguing that "AI" should not be regulated. Not because AI can't be dangerous or AI systems shouldn't be regulated, but because AI is ill-defined and a useless regulatory category.
For a special issue of Digital Government, I wrote a commentary arguing that "AI" should not be regulated. Not because AI can't be dangerous or AI systems shouldn't be regulated, but because AI is ill-defined and a useless regulatory category.
The German Informatics Society (@informatik.bsky.social) will honour me as Junior Fellow at their annual meeting in two weeks. I'm very flattered to be in the company of very talented and bright people (and to still be considered young 😉).
KI, Medizin, Recht und Diversität: GI ernennt vier Informatik-Talente zu Junior-Fellows
Vier junge Menschen, die die Informatik voranbringen: Am 24. September 2024 werden Franziska Boenisch, Dr. Daniel Braun, Bettina Finzel und Mareike Lisker von der Gesellschaft für Informatik e.V. (GI)...
gi.de
September 9, 2024 at 8:13 AM
The German Informatics Society (@informatik.bsky.social) will honour me as Junior Fellow at their annual meeting in two weeks. I'm very flattered to be in the company of very talented and bright people (and to still be considered young 😉).
I’m in Bangkok for #ACL2024 and will talk about Legal NLP:
On Monday I present our corpus and paper AGB-DE (arxiv.org/abs/2406.06809) at the third poster session.
On Thursday I will talk about teaching NLP in Law School as part of the Workshop on Teaching NLP.
AGB-DE: A Corpus for the Automated Legal Assessment of Clauses in...
Legal tasks and datasets are often used as benchmarks for the capabilities of language models. However, openly available annotated datasets are rare. In this paper, we introduce AGB-DE, a corpus...
arxiv.org
August 12, 2024 at 12:59 AM
I’m in Bangkok for #ACL2024 and will talk about Legal NLP:
On Monday I present our corpus and paper AGB-DE (arxiv.org/abs/2406.06809) at the third poster session.
On Thursday I will talk about teaching NLP in Law School as part of the Workshop on Teaching NLP.
That’s quite interesting: A report from Goldman Sachs questioning whether the use of Generative AI is actually financially sustainable. I personally think we will see more and more that it is often not, particularly for tasks where “simpler” models exist.
www.goldmansachs.com/intelligence...
www.goldmansachs.com/intelligence...
Gen AI: too much spend, too little benefit?
www.goldmansachs.com
July 2, 2024 at 9:19 PM
That’s quite interesting: A report from Goldman Sachs questioning whether the use of Generative AI is actually financially sustainable. I personally think we will see more and more that it is often not, particularly for tasks where “simpler” models exist.
www.goldmansachs.com/intelligence...
www.goldmansachs.com/intelligence...