Marcel Bollmann's Avatar

Marcel Bollmann

@marcel.bollmann.me

Associate professor at @liu.se ๐Ÿ‡ธ๐Ÿ‡ช, site development lead for @aclanthology.org, editor-in-chief at @nejlt.bsky.social. Mildly obscure #NLP researcher. I like coffee and board games. ๐Ÿ  https://marcel.bollmann.me/

1,709 Followers  |  353 Following  |  117 Posts  |  Joined: 10.11.2024  |  2.2857

Latest posts by marcel.bollmann.me on Bluesky

AI Allergy I remember being excited about AI. I remember 20 years ago, being excited about neuroevolutionary methods for learning adaptive behaviors in...

New blog post: AI Allergy.

On my increasing disgust with the AI discourse, even though I still like the technical and philosophical. And how I wish I could be excited about AI again.

togelius.blogspot.com/2025/08/ai-a...

13.08.2025 05:00 โ€” ๐Ÿ‘ 88    ๐Ÿ” 17    ๐Ÿ’ฌ 5    ๐Ÿ“Œ 5

"Hang on a moment while we sign you out." - sign me in, outlook, IN!

30.07.2025 06:39 โ€” ๐Ÿ‘ 6    ๐Ÿ” 1    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Writing on any kind of social network always feels like approaching strangers in the Wild West โ€“ they always react like you might pull a gun at them any second. I'm so tired of it.

27.07.2025 00:51 โ€” ๐Ÿ‘ 4    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Making changesโ€”meaningless or notโ€”with the sole intent of gaming the peer-review system sounds like a better way of defining misconduct here. But the hard problem in practice is of course proving the authorโ€™s intent.

25.07.2025 16:43 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

My point is that โ€œchanges to a paper not made with the intention to improve itโ€ is not a good criterion for misconduct IMO, as we make changes for other reasons all the time (including what Emile said about appeasing reviewers).

25.07.2025 16:43 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
A screenshot from the DeepL translator, showing a Swedish translation with โ€œAlternativesโ€ underneath it, which contains entirely Japanese text.

A screenshot from the DeepL translator, showing a Swedish translation with โ€œAlternativesโ€ underneath it, which contains entirely Japanese text.

DeepL when translating to Swedish: Have you considered writing in Japanese?

25.07.2025 13:55 โ€” ๐Ÿ‘ 2    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

We make meaningless changes all the time e.g. to fit within page limits, those also often don't make the paper better in the eyes of the authors...

25.07.2025 12:16 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Microsoft Forms

๐ŸŽ“ Looking for PhD opportunities in #NLProc for a start in Spring 2026?

๐Ÿ—’๏ธ Add your expression of interest to join @copenlu.bsky.social here by 20 July: forms.office.com/e/HZSmgR9nXB

Selected candidates will be invited to submit a DARA fellowship application with me: daracademy.dk/fellowship/f...

27.06.2025 06:51 โ€” ๐Ÿ‘ 15    ๐Ÿ” 13    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

a man throws a glass apple at a concrete egg, causing it to shatter

21.06.2025 18:13 โ€” ๐Ÿ‘ 61    ๐Ÿ” 6    ๐Ÿ’ฌ 3    ๐Ÿ“Œ 7

ICYMI โ€“ still a few days left to apply! โคต๏ธ

28.05.2025 13:41 โ€” ๐Ÿ‘ 4    ๐Ÿ” 3    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
LEGO website in 1996

LEGO website in 1996

MTV Online website in 1996

MTV Online website in 1996

Pepsi website in 1996

Pepsi website in 1996

WinZip website in 1996

WinZip website in 1996

Web design trends in 1996

โ€“ LEGO
โ€“ MTV Online
โ€“ Pepsi
โ€“ WinZip

#WebDesignHistory

23.05.2025 12:19 โ€” ๐Ÿ‘ 193    ๐Ÿ” 58    ๐Ÿ’ฌ 10    ๐Ÿ“Œ 14

Same here, same here

20.05.2025 09:44 โ€” ๐Ÿ‘ 0    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Post image

#NAACL2025 ended more than a week ago & @ufal-cuni.bsky.social folks were there:
Main conf: @kathaem.bsky.social presented joint work w/ @tomlim.bsky.social, @jlibovicky.bsky.social and Alex Fraser: Beyond Literal Token Overlap: Token Alignability for Multilinguality aclanthology.org/2025.naacl-s...

16.05.2025 12:07 โ€” ๐Ÿ‘ 14    ๐Ÿ” 3    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0
Stellen OBP - Georg-August-Universitรคt Gรถttingen Webseiten der Georg-August-Universitรคt Gรถttingen

Interested in multilingual tokenization in #NLP? Lisa Beinborn and I are hiring!

PhD candidate position in Gรถttingen, Germany: www.uni-goettingen.de/de/644546.ht...

PostDoc position in Leuven, Belgium:
www.kuleuven.be/personeel/jo...

Deadline 6th of June

16.05.2025 08:23 โ€” ๐Ÿ‘ 25    ๐Ÿ” 13    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 2
Preview
Postdoc in Computational Social Science

๐Ÿ“ฃ Weโ€™re looking for postdocs in computational social science at Linkรถping University, Sweden!

Positions can be based at the LiU NLP group (though the call is not limited to NLP). Two years full-time, with possibility of a third year. Deadline is June 3, 2025.

More info here: liu.se/en/work-at-l...

14.05.2025 16:13 โ€” ๐Ÿ‘ 3    ๐Ÿ” 2    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1
Preview
5 minutes with Marshall Taylor Meet Marshall A. Taylor from New Mexico State University, a computational sociologist studying how cultural knowledge changes across populations using language and survey data.

โœจ Please join us in welcoming Marshall A. Taylor to the Swedish Excellence Center for Computational Social Science (SweCSS) as our new Senior Visiting Fellow: liu.se/en/news-item...

09.05.2025 09:00 โ€” ๐Ÿ‘ 9    ๐Ÿ” 1    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0
Norrkรถping campus at sunset

Norrkรถping campus at sunset

We are hiring postdocs in Computational Social Science
๐Ÿ“SweCSS, Norrkรถping, Sweden
โฐDeadline June 3
๐Ÿ”—https://liu.se/en/work-at-liu/vacancies/26854
Please apply // help us spread the word

13.05.2025 12:47 โ€” ๐Ÿ‘ 106    ๐Ÿ” 76    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 1
Video thumbnail

Eveyone & their brother is using ChatGPT to avoid doing the reading/homework & weโ€™re all going to suffer bc of it.

07.05.2025 19:11 โ€” ๐Ÿ‘ 7828    ๐Ÿ” 1920    ๐Ÿ’ฌ 433    ๐Ÿ“Œ 212
Post image

PhD Timeline xkcd.com/3081

25.04.2025 15:32 โ€” ๐Ÿ‘ 60707    ๐Ÿ” 20851    ๐Ÿ’ฌ 611    ๐Ÿ“Œ 840

Do you have any favorite resources for โ€œhow to write feedback to studentsโ€?

In my courses, itโ€™s often several TAs that assess student papers, and I notice that the style and substance of the feedback differs wildly. I wanted to compile some guidelines, but was wondering whatโ€™s already out there.

15.04.2025 16:35 โ€” ๐Ÿ‘ 3    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

I just attended a conference in the US. We had one speaker detained and two others turned away at the border.

13.04.2025 23:10 โ€” ๐Ÿ‘ 7173    ๐Ÿ” 3353    ๐Ÿ’ฌ 129    ๐Ÿ“Œ 195

First vs. second syllable stress: /mษ™หˆnaษชษ™kli/

12.04.2025 14:39 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

I also noticed a pattern: for me itโ€™s often those where the stressed syllable changes from the more common word โ€” maniac/maniacally, demonstrate/demonstrable, carnivore/carnivorous, admire/admirable, ...

Strong accent is also often perceived as low language proficiency/understanding, I think.

12.04.2025 14:38 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

Todayโ€™s word is โ€œmaniacallyโ€.

12.04.2025 12:21 โ€” ๐Ÿ‘ 9    ๐Ÿ” 0    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 0

Back on X, I had an ongoing thread about English words of which I had *just* discovered the correct pronunciation.

It contained words like conscientious, demonstrable, carnivorous. The last one I posted in 11/2021: epitome.

I thought by now I wouldn't discover any new ones anymore. I was wrong. >

12.04.2025 12:21 โ€” ๐Ÿ‘ 10    ๐Ÿ” 0    ๐Ÿ’ฌ 2    ๐Ÿ“Œ 0
Preview
Bitrรคdande universitetslektor i datorlingvistik Uppsala universitet, Institutionen fรถr lingvistik och filologi Institutionen fรถr lingvistik och filologi bedriver forskning och utbildning inom sprรฅkvetenskap och textstudier med fokus pรฅ Asiens s

We just announced a new position as assistant professor / bitrรคdande universitetslektor (BUL).
For more information visit:
uu.varbi.com/se/what:job/...

09.04.2025 16:26 โ€” ๐Ÿ‘ 5    ๐Ÿ” 3    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1
I think you are misunderstanding what this tech demo actually is, but I will engage with what I think  your gripe is โ€” AI tooling trivializing the skillsets of programmers, artists, and designers.

My first games involved hand assembling machine code and turning graph paper characters into hex digits.  Software progress has made that work as irrelevant as chariot wheel maintenance.

Building power tools is central to all the progress in computers.

Game engines have radically expanded the range of people involved in game dev, even as they deemphasized the importance of much of my beloved system engineering.

AI tools will allow the best to reach even greater heights, while enabling smaller teams to accomplish more, and  bring in some completely new creator demographics.

Yes, we will get to a world where you can get an interactive game (or novel, or movie) out of a prompt, but there will be far better exemplars of the medium still created by dedicated teams of passionate developers.

The world will be vastly wealthier in terms of the content available at any given cost.

Will there be more or less game developer jobs? That is an open question. It could go the way of farming, where labor saving technology allow a tiny fraction of the previous workforce to satisfy everyone, or it could be like social media, where creative entrepreneurship has flourished at many different scales. Regardless, โ€œdonโ€™t use power tools because they take peopleโ€™s jobsโ€ is not a winning strategy.

I think you are misunderstanding what this tech demo actually is, but I will engage with what I think your gripe is โ€” AI tooling trivializing the skillsets of programmers, artists, and designers. My first games involved hand assembling machine code and turning graph paper characters into hex digits. Software progress has made that work as irrelevant as chariot wheel maintenance. Building power tools is central to all the progress in computers. Game engines have radically expanded the range of people involved in game dev, even as they deemphasized the importance of much of my beloved system engineering. AI tools will allow the best to reach even greater heights, while enabling smaller teams to accomplish more, and bring in some completely new creator demographics. Yes, we will get to a world where you can get an interactive game (or novel, or movie) out of a prompt, but there will be far better exemplars of the medium still created by dedicated teams of passionate developers. The world will be vastly wealthier in terms of the content available at any given cost. Will there be more or less game developer jobs? That is an open question. It could go the way of farming, where labor saving technology allow a tiny fraction of the previous workforce to satisfy everyone, or it could be like social media, where creative entrepreneurship has flourished at many different scales. Regardless, โ€œdonโ€™t use power tools because they take peopleโ€™s jobsโ€ is not a winning strategy.

John Carmack on the AI Quake-hallucinator.

08.04.2025 09:12 โ€” ๐Ÿ‘ 31    ๐Ÿ” 5    ๐Ÿ’ฌ 1    ๐Ÿ“Œ 0

๐Ÿซ 

01.04.2025 00:40 โ€” ๐Ÿ‘ 1    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 0

Iโ€™m currently forced to prepare a document in a given MS Word template, and words canโ€™t describe how hilariously terrible the experience is. Completely impossible to paste text and tables from another document without the styling being all over the place, and no obvious way to fix it.

31.03.2025 16:11 โ€” ๐Ÿ‘ 4    ๐Ÿ” 0    ๐Ÿ’ฌ 0    ๐Ÿ“Œ 1

@marcel.bollmann.me is following 20 prominent accounts