Computational Linguistics @ QMUL

Computational Linguistics Lab @ QMUL; language, meaning, social media, web searches, dialogue systems, creativity, mental health,...

[yin, wenjie @eijnewniy]

📑new paper on the generalisability of abusive language detection models and the role of slurs and profanity in training!

💭non-abusive swearing and implicit abuse are often confusing for models. _how_ do such biases come into play?

💡open access on:
https://t.co/OlVFXUwKOD

[Matthew Purver @mpurver]

Nice blog post @BBCRD by @QMCompLing 's very own Iacopo Ghinassi - read about his work in topic segmenation #NLProc

[Ravi Shekhar @RShekhar_vision]

#NLProc come work with us at @QMCompLing, @QMCompLing, we have multiple Postdoctoral openings. 2 PDRA position in @embeddiaproject and SoDeStream (https://t.co/ksoGfWVb2x) projects

[Computational Linguistics @ QMUL @QMCompLing]

Postdoc research positions in NLP, dialogue modelling, social data science with us at @QMEECS:

[Gareth Tyson @Gareth_Tyson]

We're recruiting 2x postdocs at @QMEECS to look at social decision making in the @ietf and user generated content moderation in the @embeddiaproject https://t.co/Nf1NsFwDgS @mpurver @Ignactro @Pat_Healey

[QMUL Sci & Eng News @QMULSciEng]

Queen Mary's Professor Shalom Lappin from @QMEECS comments on the case for regulating big tech in this article for @guardiantech 👇

[Games and NLP @GamesNlp]

We recently heard the great news that our ARCIDUCA proposal on conversational agents in games has been funded by @EPSRC!

#NLProc #MachineLearning #BigData #games #AI

[embeddia @embeddiaproject]

A snippet from our programme on December 8th:

@mpurver from @QMCogSci, @QMUL and Marko Pranjić from @trikoder will present a #comment #filtering and #hatespeech detection demo.

Don't forget to #register!

#embeddiaconference

https://t.co/a9fzrOK0hw

[Xia Zeng @XiaZeng2]

We (with @AmaniAbumansour and @arkaitz) have a survey paper on automated fact-checking published on Language and Linguistics Compass! Check it out here: https://t.co/1AgEXeXCKR🤩

[Computational Linguistics @ QMUL @QMCompLing]

(5/5) We then propose two approaches to improve robustness: MAS augments input data with synonyms to increase lexical variation during training, and ADV uses adversarial training. Both are pretty effective at improving the performance. See the paper for more detail:

[Computational Linguistics @ QMUL @QMCompLing]

(1/5) New ACL paper out now: Towards Robustness of Text-to-SQL Models against Synonym Substitution, Gan, Chen, Chang, Purver, Woodward, Xie, Huang: improving Text-to-SQL robustness via a new dataset, input data augmentation and adversarial training https://t.co/IWumcHpoBp #NLProc

[Computational Linguistics @ QMUL @QMCompLing]

(4/5) Now we can test how much substituting in synonymous material affects state-of-the-art models. Accuracy drops dramatically!

[Xia Zeng @XiaZeng2]

Excited to shared our binary step-by-step approach that achieves substantial improvements over the baseline!🍻 No extra data or huge models required!😆

[Arkaitz Zubiaga @arkaitz]

We have an opening for a lecturer in #NLProc to work with us at @QMEECS @QMCompLing -- apply by 20th May & help spread the word!

https://t.co/ApGx1boMoC

[Pat Healey @Pat_Healey]

Congratulations to Morteza Rohanian and Julian Hough, on their COLING 2020 Outstanding Paper: "Re-framing Incremental Deep Language Models for Dialogue Processing with Multi-task Learning"! @QMCogSci

https://t.co/UpbMOYMm8E

[Computational Linguistics @ QMUL @QMCompLing]

Congratulations to Morteza Rohanian & Julian Hough: new incremental dialogue work announced as a COLING 2020 Outstanding Paper: https://t.co/W1ndessiDf https://t.co/qVqWS9birB

[Computational Linguistics @ QMUL @QMCompLing]

Congratulations to our own Morteza Rohanian and Julian Hough for an outstanding paper award at @coling2020 🥳

https://t.co/W1ndessiDf

[Computational Linguistics @ QMUL @QMCompLing]

New #NLProc book from Prof Shalom Lappin, "Deep Learning and Linguistic Representation": https://t.co/GdZWXiJVAK

[Social Data Science @ QMUL @QMSDS]

We're guest editing 2 special issues:

OSNEM SI on Detecting, Understanding and Countering Online Harms (deadline 15th Nov)
https://t.co/8uY2nYXTxo

PUC SI on Intelligent Systems for Tackling Online Harms (deadline 4th Dec)
https://t.co/vubGRTojQ7