Good news (for me!) my gender bias paper from 2023 still replicates with GPT-5.
Bad news (for everyone!) my gender bias paper from 2023 still replicates with GPT-5.
arxiv.org/pdf/2308.14921
hkotek.com/blog/gender-...
@saad.me.uk
Natural Language Generation Expert and Lead Data Scientist at trivago. Active researcher in the field of NLG. Board member of @siggen.bsky.social. ๐ Homepage: https://saad.me.uk #NLG #NLP #NLProc.
Good news (for me!) my gender bias paper from 2023 still replicates with GPT-5.
Bad news (for everyone!) my gender bias paper from 2023 still replicates with GPT-5.
arxiv.org/pdf/2308.14921
hkotek.com/blog/gender-...
My south travel game is lackingโฆ
07.08.2025 15:08 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Furthest Iโve been:
N: Jyvรคskylรค, Finland
E: Yubari, Hokkaido, Japan
S: Muscat, Oman
W: Starved Rock State Park, IL, USA
๐ข Our Call for Nominations for ARR co-Editors-in-Chief is still open!
๐๏ธ Deadline extended to August 24, 2025.
For details and nomination instructions, visit:
๐ aclrollingreview.org/call-for-nom...
#ARR #NLProc
In the Future All Food Will Be Cooked in a Microwave, and if You Canโt Deal With That Then You Need to Get Out of the Kitchen
03.08.2025 20:35 โ ๐ 154 ๐ 90 ๐ฌ 7 ๐ 15Back to sending undergraduates in rooms full of UNIX terminals with and writing code in VIM/Emacs.
06.08.2025 02:51 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0BARCODES.
youtu.be/ng-qtQbMC9w
No one gives em a second thought anymore. Yet they were once cutting edge tech! Heck, computer magazines even had whole sections dedicated to UPC scanning & software. So I finally got a 1994 PC barcode reader kit and used it with some POS applications! ...and Duke 3D
Also looks like itโs hobbled on multilingual performance compared to its peers from the quick glance I had in the technical report.
05.08.2025 20:48 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Danke fรผr diese Initiative. Ich stimme zu, dass wir als eine Gesellschaft diesen โDritte Platzโ brauchen. Sonntagsรถffnungen geben Familien und Individuen in der Stadt noch mehr Optionen.
03.08.2025 14:30 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0#Sonntags in die Bibliothek? Unbedingt. In #Dรผsseldorf haben wir die Zentralbibliothek im KAP1 und die Stadtteilbรผchereien in Bilk, Eller und Unterbach* (*mit Bibliothekskarte) sonntags geรถffnet. Im Herbst folgen die Stadtteilbรผchereien in Wersten und Gerresheim! ๐ youtu.be/xg5PIcqks1w?...
03.08.2025 12:55 โ ๐ 10 ๐ 2 ๐ฌ 1 ๐ 0i'd rather no coffee than a bad coffee
01.08.2025 12:35 โ ๐ 31 ๐ 6 ๐ฌ 5 ๐ 2Neat year #NLP conferences:
๐ #EACL2026: 24โ29 March ๐ฒ๐ฆ Rabat (ARR deadline: 6 Oct)
๐ #LREC2026: 11โ16 May ๐ช๐ธ Palma de Mallorca (deadline: 17 Oct)
๐ #ACL2026: 2โ7 July ๐บ๐ธ San Diego (ARR deadline: 6 Jan)
๐ #COLING2027: 9โ14 May 2027 ๐ฒ๐ด Macao
Someone on here compared AI produced material as "digital asbestos " which in a few decades' time we will have to work out how to eradicate from the fabric of research.
30.07.2025 08:39 โ ๐ 3815 ๐ 1398 ๐ฌ 43 ๐ 43@shaneir.bsky.social I have only just noticed that Margaret Boden died a few days ago at age 88.
In the mid-70s, her AI books were required reading for those of us working on various UK computing research projects.
Ten years ago she gave this presentation.
www.youtube.com/watch?v=wPRA...
Would it be a bad time to advertise that we will have a wonderful NLG conference in Hanoi, Vietnam this year? ๐
2025.inlgmeeting.org
ICML found hidden prompts in accepted papers. They have released a statement icml.cc/Conferences/...
Yes, itโs unacceptable. So is using an LLM to review a paper. Peer review is so broken.
Countdown clock for the #INLG2025 direct submission deadline: www.timeanddate.com/countdown/ge...
Submit to the longest running conference focused on #NaturalLanguageGeneration :D
(ARR commitment deadline is August 7th)
๐ข ACL 2025 this is a reminder of ๐ค #InvitedTalk by Verena Rieser from Google DeepMind "Whose Gold? Re-imagining Alignment for Truly Beneficial AI." on the technical & ethical challenges of building beneficial AI! ๐ค โ๏ธ #ACL2025NLP #NLProc
Find out more: 2025.aclweb.org/program/keyn...
The first papers are already submitted, but for anyone hoping for an extra week to work on their paper, I have very good news!
The deadline for submitting to #INLG2025 has been extended until July 18th (end of the day anywhere on earth, AoE).
Submit your work on generation!
2025.inlgmeeting.org
We ran a randomized controlled trial to see how much AI coding tools speed up experienced open-source developers.
The results surprised us: Developers thought they were 20% faster with AI tools, but they were actually 19% slower when they had access to AI than when they didn't.
From a commercial research perspective:
โข Hard to convince organisations to publish the results of A/B tests due to competition.
โข Even if they can be published the standard for statistical significance is lower than academia. For example, Agoda uses p-values up to 0.4: medium.com/agoda-engine...
Ive written a "Last Word" opinion piece for CL about evaluating real-world impact. It
* looks at how impact can be evaluated
* shows via a structured survey that perhaps 0.1% of ACL Anth papers measure real-world impact
* discusses why this is the case
arxiv.org/abs/2507.05973
Sounds like something to train a spam filter on.
08.07.2025 15:52 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0A picture of a beautiful modern building against a luscious green backdrop and open blue sky. This is the VIASM building where the conference will be hosted.
Announcing 4 keynote speakers for #INLG2025!
Verena Rieser (Google DeepMind)
Hadas Kotek (Apple)
Minlie Huang (้ปๆฐ็; Tsinghua University)
Mike White (Ohio State University)
It's shaping up to be a great conference!
Don't forget to submit by 14 July AoE:
2025.inlgmeeting.org
We can also announce two of our keynote speakers: Verena Rieser (Google DeepMind) and Minlie Huang (Tsinghua University)!
If you want the latest updates on SIGGEN activities, including workshops and other events, subscribe to the mailing list!
www.jiscmail.ac.uk/cgi-bin/wa-j...
The Second #CallForPapers just went to the SIGGEN list!
Submit work on any aspect of #NaturalLanguagGeneration, including but not limited to: rule-based data-to-text systems, summarisation & simplification with the latest LLMs, or new evaluation methods :)
Deadline: 14 July
2025.inlgmeeting.org
Research on trustworthy text generation by @tuetschek.bsky.social was featured in the Czech edition of @wired.com ๐
www.wired.cz/clanky/vyvij...
When I talk to non-tech people, a lot of them don't realise that these summaries are automatically generated, nor that the tech used to generate them is so buggy. They assume this is a source of truth from Google, etc, which is a natural assumption for a search engine for a normal person.
07.06.2025 15:08 โ ๐ 6 ๐ 3 ๐ฌ 0 ๐ 0๐ Don't forget! The deadline for submitting your abstract to the #CLIN conference in Leuven is coming: 13th of June! Submitting is easy: name, title of your work, 500-word abstract, done! #nlp #nlproc #compling #llm #ai #dutch clin35.ccl.kuleuven.be
02.06.2025 07:08 โ ๐ 1 ๐ 2 ๐ฌ 0 ๐ 2