Dave in 🏴
SIGGENPapers from the recent #INLG2023 are up on the #ACLAnthology now!
Go read some of the latest papers in #NaturalLanguageGeneration, if that's your cup of tea :)
Thank you to everyone who made #INLG2023 (joint with #SIGDIAL2023) a great success! Now it's time to start thinking about #INLG2024 and #INLG2025!
If you're interested in helping with the scientific programme or volunteering to host the conference next year or the following year, please check out the #CallForBids: https://www.jiscmail.ac.uk/cgi-bin/wa-jisc.exe?A2=ind2309&L=SIGGEN&O=D&P=3591
Well this is exciting! We got the best short paper award for our paper "Building a Dual Dataset of Text- And Image-Grounded Conversations and Summarisation in Gàidhlig (Scottish Gaelic)" at #INLG2023! Work with Dimitra Gkatzia, Will Lamb, and Anna Groundwater
Mark DingemanseSince #sigdialxinlg2023 #sigdial2023 #inlg2023 are ongoing and LLMs are (still) the talk of the town, let me share our paper on the openness of instruction-tuned LLMs, raising some key concerns for anyone who cares about scientific reproducibility, ethical knowledge production, and the risks of relying on proprietary and closed solutions
Paper here (pdf): https://pure.mpg.de/rest/items/item_3526897_1/component/file_3526898/content
Live tracker here: https://opening-up-chatgpt.github.io/
Thread here: https://scholar.social/@dingemansemark/110728766949576969
Also biased and excited to hear @xudong_hong present the "Visually Grounded Story Generation Challenge" at #INLG2023 #GenChal
Very biased, but also very excited about Khyathi Chandu's presentation of our new proposed shared task at #INLG2023: "LowReCorp: The Low-Resource NLG Corpus Building Challenge"
Join the #SharedTask during the coming year if you want to use our UI or task design to collect #NLG data for #LowResourceLanguages!
#DialogueSummarization #QuestionAnswering #ResponseGeneration
Now Tom Williams is presenting "Exploring the Naturalness of Cognitive Status-Informed Referring Form Selection Models" at #INLG2023
Judith Sieker presents "Beyond the Bias: Unveiling the Quality of Implicit Causality Prompt Continuations in Language Models" at #INLG2023
Language models struggle with implicit causality verbs, so the researchers set out to study ICs and coreference with these models
In this afternoon session on evaluation and linguistic analysis at #INLG2023, we have Fahime Same, Fafa, presenting "Models of Reference Production: How Do They Withstand the Test of Time?"