Lightnews — Scholar-powered news

Florian Eichin @florian-eichin.com · Aug 13

Reposted by Florian Eichin

Sven Nyholm @svennyholm.bsky.social · Aug 7

Thank you to our great Munich Center for Machine Learning (@munichcenterml.bsky.social) for featuring me in this research film! Lots of great films clips with my MCML colleagues are available on MCML's YouTube channel. #ai #aiethics #mcml #philosophy youtu.be/KUqiY8o1yng?...

2 7

Reposted by Florian Eichin

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 27

Unsure which presentations to attend at #ACL2025? 🛎️🗣️

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 23

Headed to ACL? MaiNLP & our most recent work will be there too👥📄
Come see what we’ve been working on!

2 4

Reposted by Florian Eichin

Janet Liu @janetlauyeung.bsky.social · Jul 23

🕺🏼swing by our poster in Hall 4/5 on Wednesday, July 30 at 11:00 to chat with @florian-eichin.com and I to find out the answers to these questions

🛎️ bonus: to see the full poster 🫣🧩

#ACL2025 #NLProc

part of the poster presentation at ACL 2025

1 3

Florian Eichin @florian-eichin.com · Jul 23

is the bike doing fine though?? 😥

1

Reposted by Florian Eichin

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 23

📝Probing LLMs for Multilingual Discourse Generalization Through a Unified Label Set
🔎Do LLMs encode and generalize discourse knowledge across languages?
👥 @florian-eichin.com @janetlauyeung.bsky.social @mhedderich.bsky.social @barbaraplank.bsky.social
🔗 arxiv.org/abs/2503.10515
📁Main - Long

1 1 3

Florian Eichin @florian-eichin.com · Jul 23

Some recommendations for #ACL2025 👇
(join me and @janetlauyeung.bsky.social to talk about discourse generalization and probing!)

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 23

Headed to ACL? MaiNLP & our most recent work will be there too👥📄
Come see what we’ve been working on!

1 3

Reposted by Florian Eichin

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 23

Headed to ACL? MaiNLP & our most recent work will be there too👥📄
Come see what we’ve been working on!

1 5 14

Reposted by Florian Eichin

Naomi Saphra @nsaphra.bsky.social · Jul 20

XAI’s dogwater performance on the 2025 IMO confirms that their Grok 4 benchmark claims were hot air. Their eye popping metrics were down to the following innovations:
- train on test
- train on test
- train on test

6 38 310

Reposted by Florian Eichin

Philipp Mondorf @pmondorf.bsky.social · Jul 18

📄 [ACL 2025 main] LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks (doi.org/10.48550/arX...)

LLMs instead of Human Judges? A Large Scale Empirical Study across 20 NLP Evaluation Tasks

There is an increasing trend towards evaluating NLP models with LLMs instead of human judgments, raising questions about the validity of these evaluations, as well as their reproducibility in the case...

doi.org

1 4 9

Reposted by Florian Eichin

Philipp Mondorf @pmondorf.bsky.social · Jul 18

📄 [ACL 2025 main] Circuit compositions: Exploring Modular Structures in Transformer-Based Language Models (doi.org/10.48550/arX...)

Circuit Compositions: Exploring Modular Structures in Transformer-Based Language Models

A fundamental question in interpretability research is to what extent neural networks, particularly language models, implement reusable functions through subnetworks that can be composed to perform mo...

doi.org

1 2 5

Reposted by Florian Eichin

Michael A. Hedderich @mhedderich.bsky.social · Jul 10

Interpretability meets Discourse. Congratulations to
@florian-eichin.com to his first ACL paper 🎉

Janet Liu @janetlauyeung.bsky.social · Jul 10

🦙 how well do LLMs encode discourse knowledge? does that generalize across languages?

🛎️ in our #ACL2025 paper, we uncover fascinating trends about multilingual discourse representations!

joint work w/ @florian-eichin.com @barbaraplank.bsky.social @mhedderich.bsky.social

📄 arxiv.org/abs/2503.10515

1 4

Florian Eichin @florian-eichin.com · Jul 10

Paper alert 🛎️

Janet Liu @janetlauyeung.bsky.social · Jul 10

🦙 how well do LLMs encode discourse knowledge? does that generalize across languages?

🛎️ in our #ACL2025 paper, we uncover fascinating trends about multilingual discourse representations!

joint work w/ @florian-eichin.com @barbaraplank.bsky.social @mhedderich.bsky.social

📄 arxiv.org/abs/2503.10515

5

Reposted by Florian Eichin

Janet Liu @janetlauyeung.bsky.social · Jul 10

🦙 how well do LLMs encode discourse knowledge? does that generalize across languages?

🛎️ in our #ACL2025 paper, we uncover fascinating trends about multilingual discourse representations!

joint work w/ @florian-eichin.com @barbaraplank.bsky.social @mhedderich.bsky.social

📄 arxiv.org/abs/2503.10515

1 3 16

Reposted by Florian Eichin

Ali Modarressi @amodarressi.bsky.social · Jul 9

I’ll be at @icmlconf.bsky.social next week presenting NoLiMa!
Poster on Tue July 15, 4:30–7pm (E-2312).

Happy to grab a coffee and chat about long-context, memory, research, or just to catch up.

I’ll be in Toronto for a couple of days after the conference, let me know if you’re around!

1 2 4

Reposted by Florian Eichin

MaiNLP lab, LMU Munich @mainlp.bsky.social · Jul 9

Caught some great moments at #MCML Munich AI Day 2025 last week📍
From sharp keynotes to poster debates. Our team had the chance to show some recent work, join the conversations, and bring back plenty of food for thought🧠🗣️📊

Munich Center for Machine Learning @munichcenterml.bsky.social · Jul 8

Last week, #MCML Munich AI Day 2025 kicked off with keynotes by Julia Schnabel and Tina Eliassi-Rad, brilliantly moderated by Eva Schulz.

2 7

Reposted by Florian Eichin

Maria Antoniak @mariaa.bsky.social · Jul 8

The study is here but gated: journals.sagepub.com/doi/10.3102/...

I’d be curious how these dynamics play out in our NLP review crisis. My hunch: many conscientious volunteers might be junior women. That time comes at a cost; chasing slackers means less time for rebutting my own reviews.

Amy Diehl, Ph.D. @amydiehl.bsky.social · Jul 7

In a study of professors, women got 378 new work requests over 4 weeks vs 118 for men. Women spent more time on service, advising & teaching; men on research. Orgs should track who is taking extra duties & ensure they are rewarded and distributed fairly. www.forbes.com/sites/kimels...

Being Too Helpful At Work Can Hurt Your Career—Here’s How To Say No

Women are more likely to take on behind-the-scenes duties at work—extra tasks like onboarding or event planning—and it's hurting their careers. Here's how to say no.

www.forbes.com

6 34

Reposted by Florian Eichin

Barbara Plank @barbaraplank.bsky.social · Jul 4

Thanks for the invitation to the Freiburg Institute for Advanced Studies (FRIAS) to give this year's Hermann-Paul-Center Lecture lnkd.in/d_wUeDfY

I enjoyed the visit, the great audience, and the stay in this lovely city.
Thank you
#blackforest #freiburg #breisgau

2 10

Reposted by Florian Eichin

Alexander Hoyle @alexanderhoyle.bsky.social · Jul 3

preprint is out

bsky.app/profile/alex...

Alexander Hoyle @alexanderhoyle.bsky.social · Jul 3

Here's the preprint---basically, you can just use concept erasure (github.com/EleutherAI/c...) to remove lang id

arxiv.org/abs/2507.01234

The Medium Is Not the Message: Deconfounding Text Embeddings via Linear Concept Erasure

Embedding-based similarity metrics between text sequences can be influenced not just by the content dimensions we most care about, but can also be biased by spurious attributes like the text's source ...

arxiv.org

2 5

Florian Eichin @florian-eichin.com · Jul 3

My MSc-thesis has been turned into a paper (whose framing you will probably not enjoy) that introduces a method which can be viewed as an unsupervised solution to a similar problem. Will share later to avoid biasing review process

1

Florian Eichin @florian-eichin.com · Jul 3

Interesting! And indeed very relevant as it enables control over the similarity modeled by the embeddings. Figure 2 is really cool. Which base embeddings were used for this?

1

Florian Eichin @florian-eichin.com · Jul 2

Haha can't wait. Let's continue the discussion at ACL!

1

Florian Eichin @florian-eichin.com · Jul 2

Also, I remember your other ACL 2025 paper which shows that the LLM approach comes with problems for topic quality, too? Very interesting read arxiv.org/abs/2502.14748

Large Language Models Struggle to Describe the Haystack without Human Help: Human-in-the-loop Evaluation of Topic Models

A common use of NLP is to facilitate the understanding of large document collections, with a shift from using traditional topic models to Large Language Models. Yet the effectiveness of using LLM for ...

arxiv.org

1 1

Florian Eichin @florian-eichin.com · Jul 2

Yeah, agreed and aware of your work :) though as established above, emb+clustering has its niche in large scale analysis with factors like multilinguality. There, LDA tends to have problems and TopicGPT is too expensive.

2 1

Florian Eichin @florian-eichin.com · Jul 2

Awesome! And yes, I totally understand and agree with the scepticism towards that

1