←
Isabelle Augenstein3,119 posts
Isabelle Augenstein avatar

Isabelle Augenstein

@IAugenstein

Full Professor @CopeNLU @uni_copenhagen. Formerly @ucl_nlp, @SheffieldNLP. Explainable AI, Natural Language Processing, ML.

◷ Joined February 2012▧ 218 Photos and videos
Siddhesh Pawar avatar
⟳ Isabelle Augenstein retweeted
Siddhesh Pawar@whoSiddheshp·Nov 17

Aligning LLMs with human values has become a common post-training goal. But values are connected, and current work only checks whether the target value moved, not what else changed. 📄 New paper: Unintended Side Effects of Value Training in LLMs [Link: arxiv.org/abs/2610.29922]

Post attachment
ELLIS avatar
⟳ Isabelle Augenstein retweeted
ELLIS@ELLISforEurope·Nov 17

We are proud to be a partner in ELAI, the European Laboratory for Artificial Intelligence, the pilot for a new virtual institute for frontier AI research across 17 countries. ELAI’s activities begin on 1 Jan 2027, after the Grant Agreement signature. 🔗 bit.ly/4iLS1W5

Post attachment
Isabelle Augenstein avatar
Isabelle Augenstein@IAugenstein·Sep 28

🚨Last chance -- closing in two days: call for a postdoc position on mechanistic interpretability and AI reasoning in our @CopeNLU lab ⤵️ @DIKU_Institut @AiCentreDK #NLProc #XAI #academicjobs

📣 Hiring postdoc in Mechanistic Understanding of AI Reasoning 📣 ⏳ Application deadline: 30 Sep 2026 🏁 Start date: 1 Feb 2027 💼 Contract length: 2 years 🏰 Workplace: @AiCentreDK @DIKU_Institut @CopeNLU ℹ️ Apply here: candidate.hr-manager.net/App… #academicjobs #NLProc #XAI

Isabelle Augenstein avatar
Isabelle Augenstein@IAugenstein·Sep 16

🚨Closing in two weeks: opportunity to apply for a postdoc position on mechanistic interpretability and AI reasoning in our @CopeNLU lab ⤵️ @DIKU_Institut @AiCentreDK #NLProc #XAI #academicjobs

📣 Hiring postdoc in Mechanistic Understanding of AI Reasoning 📣 ⏳ Application deadline: 30 Sep 2026 🏁 Start date: 1 Feb 2027 💼 Contract length: 2 years 🏰 Workplace: @AiCentreDK @DIKU_Institut @CopeNLU ℹ️ Apply here: candidate.hr-manager.net/App… #academicjobs #NLProc #XAI

Siddhesh Pawar avatar
⟳ Isabelle Augenstein retweeted
Siddhesh Pawar@whoSiddheshp·Sep 15

People use LLMs to answer subjective, culturally situated questions. Most benchmarks ask whether the answer is correct, but they overlook the framing features (eg, insider positioning) 📄 New #EMNLP'26 paper: "Not What, But How: A Communicative Audit of LLM Response Framing"

Post attachment
ACL 2027 avatar
⟳ Isabelle Augenstein retweeted
ACL 2027@aclmeeting·Sep 11

ACL Sustainable Reviewing Policy: We are introducing changes in the @ReviewAcl reviewing and submissions. The changes will cap authors and introduce changes in the reviewing to keep our community sustainable. #NLProc

Sekh (Sk) Mainul Islam avatar
⟳ Isabelle Augenstein retweeted
Sekh (Sk) Mainul Islam@SKMAINULISLAM11·Sep 2

Excited to share that our work, "BiasGym: A Simple and Generalizable Framework for Analyzing and Removing Biases through Injection", was accepted at the EMNLP 2026 (Findings). 📄 Paper (Updated): lnkd.in/ejdD4s6T 💻 Code: lnkd.in/eG8FmgRY

🚀 Excited to share our new preprint: BiasGym: Fantastic LLM Biases and How to Find (and Remove) Them 📄 Read the paper: arxiv.org/abs/2508.08855

Post attachment
Sekh (Sk) Mainul Islam avatar
⟳ Isabelle Augenstein retweeted
Sekh (Sk) Mainul Islam@SKMAINULISLAM11·Sep 2

Excited to share that our work, "Multi-Step Knowledge Interaction Analysis via Rank-2 Subspace Disentanglement", was accepted at the EMNLP 2026 (Findings). Huge thanks to my amazing PhD supervisors: @IAugenstein , and @atanasovapepa ! See you all in Budapest 🇭🇺!

What is the interaction dynamics between Parametric Knowledge (PK) and Context Knowledge (CK) in generating longer Natural Language Explanation (NLE) sequences?

Greta Warren avatar
⟳ Isabelle Augenstein retweeted
Greta Warren@gretawarren_·Sep 2

How do people use images to spread misinformation online? 🖼️ We analysed 27k posts from X in 7 languages & found: 🔹Slanted framing of real images is more common than deepfakes and doctored images 🔹Vaccine misinfo borrows credibility via news screenshots arxiv.org/abs/2608.29681

Post attachment
CopeNLU avatar
⟳ Isabelle Augenstein retweeted
CopeNLU@CopeNLU·Aug 18

Last week, @atanasovapepa was awarded a HC Ørsted Research Talent Award! 🏆 The prize recognises her research on how AI and LLMs can be made more trustworthy, transparent and accountable. Huge congrats to Pepa from us at CopeNLU, so well deserved! 🎉 hcoersted.dk/h-c-oersted-pri…

Post attachment