Good news! #NLProc methods cannot detect deception! @aswathyve.bsky.social's paper got accepted to the Computational Linguistics Journal.
"What if Deception Cannot be Detected? A Cross-Linguistic Study on the Limits of Deception Detection from Text"
Preprint arxiv.org/abs/2505.13147
Interested in real world probability (estimates) and how LLMs and humans misinterpret them? Want to do a PhD in #NLProc? Please get in touch and apply in this awesome project lead by @seanpapay.bsky.social (as part of @bamnlp.de at @uni-bamberg.de)
The BamNLP Group at the University of Bamberg has the following papers and contributions at The 19th Conference of the European Chapter of the Association for Computational Linguistics (EACL 2026) in Rabat, Morocco. Moiz Rauf, Sean Papay, Georg Brosinsky, Ira Stoll: Medical Summarization in Practice: Design, Deployment, and Analysis of a Clinical Summarization System for a German Hospital (Industry Track). Yanran Chen, Lynn Greschner, Roman Klinger, Michael Klenk, Steffen Eger: Emotionally Charged, Logically Blurred: AI-driven Emotional Framing Impairs Human Fallacy Detection (Main Conference). Johannes Schäfer, Janne Wagner, Roman Klinger. Appraisal trajectories in narratives reveal distinct patterns of emotion evocation. (WASSA Workshop) Sabine Weber, Lynn Greschner, Roman Klinger. Says who? argument convincingess and reader stance are correlated with perceived author personality. (WASSA Workshop) In addition, BamNLP is involved with the conference organization: • Roman Klinger co-organizes the 5th Workshop on Computational Approaches to Subjectivity, Sentiment & Social Media Analysis, together with Jeremy Barnes, Valentin Barriere, Orphée De Clercq, Debora Nozza, Célia Nouri, and Pranaydeep Singh. • Sabine Weber is one of the Diversity and Inclusion Chairs. • Lynn Greschner and Roman Klinger propose a "Birds-of-a-Feather" session on the topic of Emotion Analysis and Generation.
We would like to share with you the contributions that we make to the @eaclmeeting.bsky.social in Rabat in March. See you all there to talk about #NLProc!
We wish you a relaxing holiday season! (and no reviewing obligations until next year) 🎄
Thanks again for your visit, the presentation, and all the interesting discussions we had!
Tomorrow I'll give my inaugural lecture @uni-bamberg.de – if you don't know what language technology #nlproc is, and want to know what @bamnlp.de works on, please feel invited! If you can't make it at 6pm to Weberei 5 in Bamberg ("Erba Campus"), send me a message and I'll share a streaming link.
photo of Barbara Plank in front of a projection of the time line how human label variation research developed in the last decade
.@barbaraplank.bsky.social visited @uni-bamberg.de to present her perspective on #labelvariation in natural language processing. A very inspiring talk, from which our #NLProc groups @bamnlp.de and @bamdial.bsky.social as well as groups working on other modalities learned a lot. Thanks, Barbara!
Happening now: Sean Papay presents his work on CRFs for distant interactions in X5, 223 :-). #xllm #nlproc #acl2025
The @uni-bamberg.de group at #acl2025 (including @bamnlp.de and @bamdial.bsky.social) #acl2025 #nlproc
Now happening in X5, 95: Johannes Schäfer presenting our work on default demographics in LLMs. #nlproc #acl2025
Now happening: Jiahui Li presents her SRW poster in X5, 218, from the entrance in the left corner. #acl2025 #nlproc
Christopher Bagdon now presents his work on multimodal social media #emotion corpus creation. hall X4, far away from the entrance, stand 375.
photo of a seminar room from the back to the front. many people sit at desks and listen to a presentation. in the front Saif Mohammad gives a talk. There are two projections next to each other of his slides.
On his way to #ACL2025, Saif Mohammad (x.com/SaifMMohammad) made a stop @bamnlp.de @uni-bamberg.de to give a very inspiring talk on "NLP for Affective Sciences".
A recording of the talk is available at youtu.be/LYC6gDJMwac
Thanks for visiting us! #NLP #NLProc
Johannes Schäfer, Aidan Combs, Christopher Bagdon, Jiahui Li, Nadine Probol, Lynn Greschner, Sean Papay, Yarik Menchaca Resendiz, Aswathy Velutharambath, Amelie Wührl, Sabine Weber, and Roman Klinger. Which demographics do LLMs default to during annotation? ACL Main Conference: Poster Session 3, Tuesday, July 29; 16:00–17:30; Hall 4/5. https://aclanthology.org/2025.acl-long.848/ Christopher Bagdon, Aidan Combs, Carina Silberer, and Roman Klinger. Donate or create? Comparing data collection strategies for emotion-labeled multimodal social media posts. ACL Main Conference: Poster Session 1, Monday, July 28; 11:00–12:30; Hall 4/5. https://aclanthology.org/2025.acl-long.847/ Jiahui Li, and Roman Klinger. iPrOp: Interactive Prompt Optimization for Large Language Models with a Human in the Loop Student Research Workshop: Tuesday, July 29, 10:30–12:00, Hall 4/5. https://aclanthology.org/2025.acl-srw.18/ Lynn Greschner, Amelie Wührl, and Roman Klinger. QoLAS: a Reddit corpus of health-related quality of life aspects of mental disorders. Workshop on Biomedical Language Processing (BioNLP): Friday, August 1, 2025; 16:00–18:00; Room 2.15. https://aclanthology.org/2025.bionlp-1.18/ Sean Papay, Roman Klinger, and Sebastian Pado. Regular-pattern-sensitive CRFs for distant label interactions. Workshop on Large Language Models and Structure Modeling (XLLM): Friday, August 1, 2025; 12:00–12:30 or 17:00–17:30; Room 1.61-62. https://aclanthology.org/2025.xllm-1.4/ Jan Hofmann, Cornelia Sindermann, and Roman Klinger. Prompt-based personality profiling: Reinforcement learning for relevance filtering. Workshop for Research on Agent Language Models (REALM): Thursday, July 31, 2025; 10:30– 11:00 or 15:00–16:00; Room 1.61-62. https://aclanthology.org/2025.realm-1.1/ Sabine Weber. Queer in AI Meet-Up. Affinity Group Meeting: Monday, July 28, 12:30–14:00; Room 1.33
We will be at @aclmeeting.bsky.social in #Vienna next week! with some main and workshop papers, and an affinity group meeting. If you want to talk to us, please have a look at the group's schedule. We are looking forward to talking to you, please don't hesitate to get in touch! #NLProc #NLP #ACL2025
Last week, we've been on a retreat to talk about the big picture of our research, research ideas, and synergies between group members – and did some team building.
@annefriedrich.bsky.social
was our external guest and helped us out of a couple of local optima. Thank you for participating!
Organized with involvement from our group 🤩 – thanks @aswathyve.bsky.social for taking the lead!
Today, we had Goran Glavaš from @uni-wuerzburg.de in our research seminar for a talk "On Multilingual Abilities of (Different Types of) Language Models" at @bamnlp.de @uni-bamberg.de Thanks for visiting us, Goran!
The BamNLP Group at the University of Bamberg has the following five papers accepted at The 63rd Annual Meeting of the Association for Computational Linguistics (ACL 2025) in Vienna. Johannes Schäfer, Aidan Combs, Christopher Bagdon, Jiahui Li, Nadine Probol, Lynn Greschner, Sean Papay, Yarik Menchaca Resendiz, Aswathy Velutharambath, Amelie Wührl, Sabine Weber, and Roman Klinger. Which demographics do LLMs default to during annotation? ACL Main Conference Christopher Bagdon, Aidan Combs, Carina Silberer, and Roman Klinger. Donate or create? comparing data collection strategies for emotion-labeled multimodal social media posts. ACL Main Conference Lynn Greschner, Amelie Wührl, and Roman Klinger. QOLAS: a Reddit corpus of health-related quality of life aspects of mental disorders. Workshop on Biomedical Language Processing (BioNLP) Sean Papay, Roman Klinger, and Sebastian Pado. Regular-pattern-sensitive CRFs for distant label interactions. Joint Workshop on Large Language Models and Structure Modeling (XLLM) Jan Hofmann, Cornelia Sindermann, and Roman Klinger. Prompt-based personality profiling: Reinforcement learning for relevance filtering. Proceedings of the First Workshop for Research on Agent Language Models (REALM)
We have five #nlproc papers accepted at @aclmeeting.bsky.social – after just one year in @uni-bamberg.de . Hope to meeting many of you in Vienna!
We are really happy about this new collaboration, in which we use our experience with #LLM models for automatic text simplification in #psychology research.
A group of three people is posing in front of a research poster titled "Fundamentals of Natural Language Processing." The poster contains sections on research focus, research methods, and possible thesis topics. The individuals are standing in an indoor academic or conference setting with other attendees visible in the background. One person is wearing a casual t-shirt with cartoon illustrations, another is in a button-up shirt, and the third is dressed in a dark sweater. The setting includes modern lighting and furniture, with a red couch and a yellow chair visible on the left side of the image.
Choose-a-chair fair at the computer science faculty @uni-bamberg.de – we presented our courses and theses topics to our students (with Sean Papay, @multilingual-s.bsky.social and @romanklinger.de).
Are you at #COLING2025? Come visit our poster presented by Yarik Menchaca Resendiz on "MOPO: Multi-Objective Prompt Optimization for Affective Text Generation". (Fri, Jan 24 13:30-15:00) #NLP #nlproc #prompts #llm #emotions
An 8 times 3 grid of photos of the members of the research group. In each photo, people wear christmas outfit.
The @bamnlp.de group wishes everybody nice holidays. 🎄
Have a look at our new #preprint on our work on human-in-the-loop prompt optimization. It's a preprint of a demo paper describing our experiment environment that we use in the @dfgpublic.bsky.social funded project INPROMPT. #nlp #nlproc
New #preprint by @nlp-amelie.bsky.social and @romanklinger.de on automatically optimizing a claim paraphrased/extractor with preference optimization, such that the claims can be better fact-checked. #nlp #nlproc
We will share the preprint as soon as it's ready. (3/3)
This leads to more robust prompts, because we can, a posteriori, decide for which objective a prompt should work, without retraining. We focus our evaluation on affective text generation across multiple domains, but it should work across other multi-objective use cases (2/3).
We got one paper accepted to #COLING2025: "MOPO: Multi-objective prompt optimization for affective text generation." with @romanklinger.de and Yarik Menchaca Resendiz. We optimize prompts for multiple objectives, while indirectly optimizing the prompts that optimize the prompts #NLP #NLProc (1/3)
