Manchester NLP

New method for robust and efficient LLM-based NLG evaluation An inversion learning method that learns effective reverse mappings from model outputs back to their input instructions, enabling the automatic generation of highly effective, model-specific evaluation prompts.

Delighted to share the #ACL2025 and TACL papers (10 in total) from the @manchester_nlp group! Come chat with our staff and students in Vienna! 🇦🇹

Happy to announce our journal paper on tongue twisters, Train and Constrain (TwistList 2.0), has now been officially published in @CompLingJournal! (Thanks to @chenghua_lin and Chen Tang) direct.mit.edu/coli/article/5… @sltcdt #nlp #nlproc #nlg

#HealTAC2025 has kicked off with a full house at the DATAMIND workshop on "Mental Health & the New NLP". #HealthTech #LLMs #MentalHealth #AIethics #datasaveslives

🎶 Don’t miss our Workshop on LLMs for Music & Audio (LLM4Music) at #ISMIR2025 in Korea 🇰🇷! Check out the submission details below👇

Prof. Steffen Eger (@egere14) from UTN gave a fantastic talk on "Literary Translation in the Era of LLMs" at our NLP group seminar ✏️📖

If you are attending #NAACL2025 in Albuquerque, check out the papers from @manchester_nlp !🤠🍻🍹

[1/n] Delighted to share our new work "COIG-P: A High-Quality and Large-Scale Chinese Preference Dataset for Alignment with Human Values". Paper: arxiv.org/abs/2504.05535 HF Daily Paper: huggingface.co/papers/2504.05… Code: github.com/multimodal-art… Data: huggingface.co/collections/m-…

1/n Delighted to share the release of our new preprint "ContrastScore: Towards Higher Quality, Less Biased, More Efficient Evaluation Metrics with Contrastive Evaluation" 📄Paper: arxiv.org/abs/2504.02106 💻Code: github.com/NL2G/contrasti…

Updates on #HEALTAC2025: submission deadline extended to April 4th: healtac2025.github.io #datasaveslives #NLProc

An absolute pleasure to host Prof. Neil Lawrence (@lawrennd) for a fascinating (and mind-bending) departmental seminar on “Information Engines: Exploring Connections Between Intelligence and Thermodynamics” — where entropy meets intelligence

Co-organised by @Manchester_NLP , @durham_uni, @PittTweet, @hkbaptistu, @SheffieldNLP, and @Yale

NLG coming to Vietnam and Southeast Asia for the very first time – really looking forward to the event 🎉🇻🇳

1/n Excited to announce our new paper "Does Table Source Matter? Improving Scientific Multimodal Table Understanding and Reasoning Ability"! arxiv.org/abs/2501.13042

Check out our survey on LLM-assisted scientific discovery, experimentation, and evaluation arxiv.org/abs/2502.05151

We are really delighted to have hosted @_joestacey_ from Imperial, who gave a talk on "Atom-based Approaches to Creating Interpretable NLI Models" at the @Manchester_NLP Seminar on Chinese New Year's Day 🎊🍻🎋

Give a try to our demo- it’s awesome :)