Research Fellow at Dublin City University.
Ireland
Joined August 2017
- Tweets461
- Following150
- Followers185
- Likes1.3K
Craig Thomson retweeted
Next in the @DCU-NLG group #INLG2004 line up: our amazing @HuidromRudali kicks off the first oral session of the main @inlgmeeting conference presentating her work on Differences in Semantic Errors Made by Different Types of Data-to-text Systems😍 aclanthology.org/2024.inlg-m…
Craig Thomson retweeted
Come join us for the #INLG2024 Tutorial on Human Evaluation of #NLProc System Quality on 24 Sep in Tokyo🗼🇯🇵!! Programme and summary paper available👇@JoaoSedoc @HuidromRudali @ThomsonSoftware @simon_mille @AdaptCentre @dcucomputing @inlgmeeting human-evaluation-tutorial.gi…
Craig Thomson retweeted
Very pleased to say we're running a Tutorial on Human Evaluation of #NLProc System Quality at #INLG2024 @inlgmeeting in September in Tokyo!! With @simon_mille @ThomsonSoftware @JoaoSedoc @HuidromRudali #DCUNLP 1/2
Craig Thomson retweeted
Good news everybody! You have an extra 7 days to prepare you submissions for #INLG2024! Please let everyone you know know!
The new deadline for Friday, 7th of June 11:59 PM AOE!
inlg2024.github.io/calls.htm…
Craig Thomson retweeted
Preprint of our CL paper is out reporting #ReproHum project results on Common Flaws in Running Human Evaluation Experiments in NLP. Including the 12 fun flaws in 6 non-cherrypicked experiments below. With @ThomsonSoftware and @EhudReiter. direct.mit.edu/coli/article-…
Craig Thomson retweeted
Our #ACL2023NLP paper, “Are Experts Needed? On Human Evaluation of Counselling Reflection Generation” is now online!
Find it at: aclanthology.org/2023.acl-lo…
(1/6)
Craig Thomson retweeted
#ReproNLP Shared Task on Reproducibility of Evaluations in #NLProc call for participation out 👇 repronlp.github.io/
Craig Thomson retweeted
Every #NLProc PhD student should at least once carry out a reproduction study. There’s no better way to really feel what it takes to carry out a repeatable study.
(Senior authors who haven’t done this should also try.)
Craig Thomson retweeted
Aberdeen CS is looking to hire a lecturer (junior faculty, like US assistant prof but without the tenure headaches) in computational linguistics and NLP. If anyone is interested, Im happy to answer any questions
Aberdeen CS is hiring new lecturers in Cybersecurity, Computational Linguistics, and Computer Science!
Closing date is 12 May
abdnjobs.co.uk/vacancy/lectu…
abdnjobs.co.uk/vacancy/lectu…
abdnjobs.co.uk/vacancy/lectu…
Craig Thomson retweeted
Our new #NLProc paper is finally out! A meta-scientific paper discussing barriers and enabling factors for researchers to carry out error analyses.
It's a mix of a survey study (i.e. using a questionnaire) and a position paper, with many footnotes and references.
➡️ Article published:
Barriers and enabling factors for error analysis in NLG research
-- Emiel van Miltenburg, Miruna Clinciu, Ondřej Dušek, Dimitra Gkatzia, Stephanie Inglis, Leo Leppänen, Saad Mahamood, Stephanie Schoch, Craig Thomson, Luou Wen
nejlt.ep.liu.se/article/view…
#nlp
Craig Thomson retweeted
I'm looking for solid experimental papers which evaluate effectiveness of zero/few shot prompting on realistic data-to-text tasks (eg weather or sports, not WebNLG or E2E). Suggestions welcome, esp if high-quality eval (not BLEU!). Thanks
Craig Thomson retweeted
The situation with reviewers of AI / ML papers demanding novelty has reached high farce
Something needs to be done to stop this tyranny asap!
If you have ever been found guilty of 'lack of novelty' (or are a serial offender, like me) read on. I have the solution:
🧵🧵👋🧵
Craig Thomson retweeted
✨I’m excited to advertise my fully-funded PhD position @ABDNCompSci in Adaptive NLG, co-supervised by @EhudReiter. I’m keen to recruit a motivated and creatively-minded student. If you are or know one, please take a look or share: tinyurl.com/adaptNLG. Apply by Jan 31st 2023
Craig Thomson retweeted
Many important tasks in #NLProc are avoided because they are labor intensive, costly, or time-consuming. E.g. annotating new datasets, or carrying out a manual error analysis.
How can we get researchers to do those things anyway? Or how do you do those things efficiently?
Craig Thomson retweeted
Siri or Skynet? How to separate AI fact from fiction theguardian.com/technology/2…
Craig Thomson retweeted
In case you missed our flier at #ACL2022nlp - nb this is for everyone in #NLProc and #ML not just people working on evaluation!! forms.gle/eQEkqagmEXJMq2cx7
Craig Thomson retweeted
Also note that there won't be a deadline extension for @inlgmeeting this year because reviewing proceeds through @ReviewAcl.
2 weeks to go for the Call for Papers deadline!
Anywhere on earth (UTC-12)
- March 15 2022
Submissions will be through ARR @ReviewAcl
Please ensure registration on openreview.net/ for all paper authors
Craig Thomson retweeted
If you are an #NLProc researcher working on Natural Language Generation, please consider filling in our survey: tilburghumanities.eu.qualtri…
Our goal is to learn more about barriers & enabling factors to carry out an error analysis.
(The survey was approved by the IRB at @TilburgU.)