Skip to main content
Menu

OxNLP events

OxNLP events

OXNLP 12 10 22 57 20 banner

This term, we plan to meet in weeks 1 to 8.

Location: OeRC Teaching Room (Room 277), Oxford e-Research Centre, 7 Keble Road.

Time: Fridays 4pm to 5:30pm

  • Online: Our meetings take place on zoom. We share the link in the weekly invite for the talks. But if you feel you have missed the link and want to join remotely, email Dragos, Fangru and Ved and they shall be able to provide you with a link.
  • Accessing the building: The building access is controlled by the university card. If you do not have access, knock on the door at 7 Keble Road and the receptionist should be able to let you in. Alternatively, email Dragos, Fangru or Ved when you are outside and they shall be able to let you in.
  • External Speakers: We plan to host multiple external speakers during the term. If you would like to suggest names for the speakers, please let the organizing committee know.
  • Socials: Based on the feedback at the end of Trinity 2023, we plan to organize a few socials where members can discuss their work with each other in a more informal setting. Our first social is planned for week one, Friday 13th. We shall release details on the mail list.
  • Speaking at the event: We invite student members or early researchers to present at the meetings. Based on the feedback at the end of Trinity 2023, the members felt that the types of talks that had the best impact were those that discussed some important and impactful papers in the field that most people attending could take away from. If you are interested in presenting and want to discuss more, please get in touch with the organizing committee.

See our Talk abstracts page for details of past events.

Trinity Term 2024

Week Date Topic Presenter(s)
Week 5 24/5/24

How to Know What Language Models Know

Jennifer Hu (Harvard University)

Week 4 17/5/24

Predicting and Controlling Mechanisms Learned by Large Language Models

Jack Merullo (Brown University)

Week 3 10/5/24

CAMEL: Communicative Agents for “Mind” Exploration of Large Language Model Society

Guohao Li (University of Oxford)

Week 2 3/5/24

Student flash talks

Anna George, Daniel Gottlich, Jabez Magomere, Lea Krause

Week 1 26/4/24

Leoākiko: Automatic Speech Recognition in Hawaiian

Oiwi Parker Jones

 

Hilary Term 2024

Week Date Topic Presenter(s)
Week 8 8/3/24

Panel discussion, with a theme of Benchmarking Large Language Models

Panel includes Fazl Barez, Emanuele La Malfa

Week 7 1/3/24

Truth Conditions at Scale, and Beyond

Dr Guy Emerson (University of Cambridge)

Week 6 23/2/24

From Word Models to World Models: Translating from Natural Language to the Probabilistic Language of Thought

Lionel Wong (PhD Student, MIT Brain and Cognitive Sciences)

Week 5 16/2/24

Highlights of NeurIPS 2023

Daniella Ye, Hunar Batra

Week 4 9/2/24

Papers:

  • Graph-enhanced Large Language Models in Asynchronous Plan Reasoning
  • Code Simulation Challenges for Large Language Models 

Fangru Lin, Emanuele La Malfa

Week 3 2/2/24

Highlights of EMNLP 2023

Isabelle Lorge, Myeongjun Jang

Week 2 26/1/24

Helpful or Harmful? Evaluating the Safety of Large Language Models

Paul Röttger (Postdoctoral Researcher, MilaNLP Lab at Bocconi University)

Week 1 19/1/24

Who decides how LLMs behave? Re-thinking the role of pluralistic human preferences in steering LLMs

Hannah Kirk (DPhil student,
University of Oxford)

 

Michaelmas Term 2023

Week Date Topic Presenter(s)
Week 8 1/12/23

Understanding Language Models with Formal Language Theory: Recurrent Neural Language Models as Recognizers of (Probabilistic) Formal Languages

Anej Svete (ETH Zurich)

Week 7 24/11/23

Whose LLM? Representation, bias, and applications to misinformation

Scott Hale (OII, University of Oxford)

Week 6 17/11/23

Guidance — constrained sampling with formal grammars for controlling language models

Harsha Nori (Microsoft)

Week 5 10/11/23

Sparks of Artificial General Intelligence: Early experiments with GPT-4

Ved Mathai

Week 4 3/11/23

Interpretable architectures and guided attention for neural language models

Marek Rei (Imperial College, London)

Week 3 27/10/23

When do Prompting and Prefix-Tuning Work? A Theory of Capabilities and Limitations

Aleksandar Petrov (Autonomous Intelligent Machines and Systems CDT, University of Oxford)

Week 2 20/10/23

Mismatches between human language processing and NLP

Janet Pierrehumbert

Week 1 13/10/23

Flash Talks by Members

Group members

 

Trinity Term 2023

Week Date Topic Presenter(s)
Week 8 16/6/23

When Does Aggregating Multiple Skills with Multi-Task Learning Work? A Case Study in Financial NLP

Jingwei Ni. (ETH & UZH)
Week 7 9/6/23

Narrative detection & analysis 

Ved Matthai , Felix Drinkall, Anna George & Dragos Gorduza (University of Oxford)
Week 6 2/6/23

Dynamics of meaning and knowledge growth in scientific disciplines

Peter Persoon (Oxford Martin School)
Week 5 26/5/23

Natural Language Processing and Mental Health

Prof Pushpak Bhattacharyya (IIT Bombay)
Week 4 19/5/2023 On bias, trustworthiness, and safety of language models William Wang (UCSB)
Week 3 12/5/2023

Modular Deep Learning

Edoardo Ponti (University of Edinburgh)
Week 2 5/5/2023

Faking Fake News for Real Fake News Detection: Propaganda-loaded Training Data Generation

Ved Mathai
Week 1 28/4/2023 Discussion Isabelle Lorge (Discussion leader)

 

Hilary Term 2023

Week Date Topic Presenter(s)
Week 8 6/3/2023 A real-time platform for contextualized conspiracy theory analysis Calvin Cheng
Week 7 27/2/2023 Is ChatGPT a General-Purpose Natural Language Processing Task Solver?

Fangru Lin

Week 6 20/2/2023 Is Neuro-Symbolic AI Meeting its Promise in Natural Language Processing? A Structured Review Ved Mathai
Week 5 13/2/2023 Guest speaker talk: NLP for Economics Elliott Ash (ETH Zurich)
Week 4 6/2/2023 Dissociating language and thought in large language models: a cognitive perspective Janet Pierrehumbert, Felix Drinkall, Valentin Hofmann
Week 3 30/1/2023 Time Aware Language Models as Temporal Knowledge Bases Fernando Almansa
Week 2 23/1/2023 CICERO - Human-level play in the game of Diplomacy by combining language models with strategic reasoning Dragos Gorduza (OMI)
Week 1 16/1/2023 Review of EMNLP papers Paul Röttger (OII), Hannah Kirk (OII), Valentin Hofmann (OeRC)
Week 0 9/1/2023 Organizational meeting Janet Pierrehumbert

 

Michaelmas Term 2022

Week Date Topic and Links to Papers Presenter(s)
MT Week 8 28/11/2022 Evaluating the Generalisability of Neural Rumour Verification Models Elena Kochkina (Queen Mary University, London)
MT Week 7 21/11/2022 Challenges in NLP for Analyzing Social Media during Emerging Events" Anjalie Field (Stanford University)
MT Week 6 14/11/2022 On emergent linguistic structures in NLP
(1) https://www.pnas.org/doi/10.1073/pnas.1907367117
(2) https://arxiv.org/pdf/2210.17406.pdf
(3) https://arxiv.org/pdf/2104.06644.pdf
Emanuele La Malfa
(Oxford CS)
MT Week 5 7/11/2022 Should Semantic Vector Composition Be Explicit? Can it be Linear?
https://aclanthology.org/2021.semspace-1.8.pdf
Isabelle Lorge
(OeRC)
MT Week 4 31/10/2022 Detecting employment status and detecting depression on Twitter
https://aclanthology.org/2022.acl-long.453.pdf
Manuel Tonneau
(OII)
MT Week 3 24/10/2022 Quick-fire talks: Past research and future research agenda by group members (5 min each) Felix Drinkall, Emanuele La Malfa, Schyan Zafar, Max Arhens, Isabelle Lorge, Paul Rottger, Hannah, Lisa Chernenko, Fangru Lin, Calvin Cheng
MT Week 2 17/10/2022 InfoSurgeon: Cross-Media Fine-grained Information Consistency Checking for Fake News Detection
https://aclanthology.org/2021.acl-long.133.pdf#page12
Dragos Gorduza
(OMI)
MT Week 1 10/10/2022 Organizational Meeting Janet Pierrehumbert
(OeRC)
Banner artwork credit: Shirley Steele https://www.steelestudio.com