Koel Dutta Chowdhury

NLP Researcher @LstSaar | Saarland University

[Koel Dutta Chowdhury @KoelDC]

@mariusmosbach @DFKI Many congrats, best news!

[Yasaman Bahri @yasamanbb]

In our new preprint, we explain how some salient features of representational geometry in language modeling originate from a single principle - translation symmetry in the statistics of data.

https://t.co/UPgLZTJ7i6

With @dhruvakarkada, @DanKorchinski, Andres Nava, and @MatthieuWyart.

[Konstantin Dobler @konstantdobler]

Token Distillation is accepted at ICLR 2026! 🎉

[anshuman @athleticKoder]

https://t.co/lfdgFwPKKJ

[Gurusha Juneja (✈️ ICML'26) @GurushaJuneja]

There were hallucinated references at #NeurIPS2025 & @iclr_conf this year, so I built harcx https://t.co/GDi8brDvad.

A Python package to verify BibTeX citations against real academic databases. It supports papers, books, and
URLs.

Usage:
pip install harcx
harcx references.bib

[Vilém Zouhar @zouharvi]

Dissatisfied with EACL paper decisions? Fret not and submit your paper with ARR reviews to Multilingual Multicultural Evaluation workshop at EACL (both archival or nonarchival) until January 5th. 🔍🙂

https://t.co/9jHwS61B6d

[Koel Dutta Chowdhury @KoelDC]

@gsarti_ @univgroningen @GroNlp @AriannaBisazza @MalvinaNissim @gchrupala Congratulations!

[Negar Foroutan @negarforoutan]

1/ 🌍 How does mixing data from hundreds of languages affect LLM training?
In our new paper "Revisiting Multilingual Data Mixtures in Language Model Pretraining" we revisit core assumptions about multilinguality using 1.1B-3B models trained on up to 400 languages.
🧵👇

[Koel Dutta Chowdhury @KoelDC]

@linguist_cat Congrats!

[Koel Dutta Chowdhury @KoelDC]

@hsanchaita @mbodhisattwa lovely! now Paris, please.

[Yihuai Hong @YihuaiH91773]

🚀The first-ever parametric LLM Unlearning Benchmark!
We find current unlearning only modify model’s behavior without truly erasing encoded knowledge in parameters, presenting ConceptVectors Benchmark, with each vector strongly tied to a specific concept.🔗https://t.co/LYnyNXnHye

[Ferenc Huszár @fhuszar]

Maximalentropiensprachenkunst.

[Koel Dutta Chowdhury @KoelDC]

@OrtliebStefania @LstSaar Congratulations! We have to celebrate this soon. 🥂🍾🎉I'll be travelling on Tuesday, but I'm in for July 5th! 🙂

[LST @ Saarland University @LstSaar]

📜 In-Context Learning Workshop organized by Paloma García de Herreros García, Israel A. Azime and Miaoran Zhang on June 12th from 2 pm to 5 pm.
More information 👇
https://t.co/h04FhRckmj

[Niyati Bafna @ACL2026 @BafnaNiyati]

Drop by my talk at LREC-COLING on Thursday on unsupervised cognate induction between closely related data-imbalanced language pairs :) https://t.co/if2y7n1nRV

[LST @ Saarland University @LstSaar]

📢Our department LST in cooperation with DFKI is inviting applications for the W3 Professorship in Language Technology.

https://t.co/WWZrlFghUc

Please spread the word.

[Javier Ferrando @javifer_96]

[1/4] Introducing “A Primer on the Inner Workings of Transformer-based Language Models”, a comprehensive survey on interpretability methods and the findings into the functioning of language models they have led to.

ArXiv: https://t.co/UBh2ZLmYkr

[Bodhisattwa Majumder @mbodhisattwa]

Incredibly proud of our teamwork, now in @icmlconf! This position starts a series of work on data-driven scientific discovery w generative models.
Follow-ups coming soon on benchmarks, systems, & accessibility in science!
https://t.co/FzqbJQm8F4
#ICML2024 @allen_ai @ai2_aristo

[Koel Dutta Chowdhury @KoelDC]

@Swarooprm7 consider visiting Hallstatt!

[Sanchaita Hazra @hsanchaita]

🌻 Super excited about my first Computer Science publication at @naaclmeeting (main)! @mbodhisattwa and I study the language of deception and how language models fare at detecting them. And guess what we've found: https://t.co/wQLHS5F1XZ
(1/n) 🧵
@EconUofU @allen_ai