MAXIMILIAN AHRENS
Postdoctoral Research Fellow in NLP | Co-Investigator for the Alan Turing Institute Research Grant on Narrative Detection
NLP for narrative and misinformation detection | NLP for economic and financial modelling
Oxford University | Department of Engineering Science | Department of Economics | OxNLP Group | Oxford-Man Institute | Alan Turing Institute
About me
I'm a Postdoctoral Research Fellow at the University of Oxford. My main research interests lie in machine learning methods for natural language processing (NLP) and applications for narrative and misinformation detection, financial market modelling, and economic forecasting.
I'm also the Co-Principal Investigator for the Alan Turing Institute Research Grant on Narrative Detection. In this capacity, I run the NarrativeLab together with my co-investigator Michael McMahon.
My PhD thesis is on multimodal language models for economics and finance. I'm a research member and co-organiser of the OxNLP Group at Oxford that focusses on NLP for social data science applications. I'm also a research member of the Machine Learning Research Group and the Oxford-Man Institute of Quantitative Finance.
I'm the founder and chair of the Conference on NLP for Social Data Sciences (NLP SoDaS).
Prior to my PhD work, I obtained an MPhil in Economics from the University of Oxford and an M.Sc. in Specialised Economic Analysis from the Barcelona School of Economics. I previously worked with McKinsey & Company, the European Central Bank, and Man AHL.
Recent News
July 2023: I'm presenting our paper on multimodal NLP frameworks for central bank communication analysis at The Bank of England's Research Seminar Series.
July 2023: I'll be speaking at a panel on LLMs, privacy, and data security at the Eyes-Off Data Summit in Dublin.
July 2023: I'm presenting our paper on multimodal NLP frameworks for central bank communication analysis at the the 2023 Annual Meeting of the Central Bank Research Association at the Federal Reserve Bank of New York and Columbia University (New York City, USA)
June 2023: We are hosting the second edition of our annual research conference on NLP for Social Data Science (NLP SoDaS) in Oxford - 28-29 June. I'm the chair of the event. Contact me for further information.
June 2023: We published our new working paper on multimodal NLP frameworks for central bank communication analysis: "Mind Your Language: Market Reactions to Central Bank Speeches". I've also been interviewed on this research paper by the Man Group.
May 2023: We are presenting our paper on multimodal NLP frameworks for central bank communication analysis at the ECONDAT 2023 Conference for Economics with non-traditional data and analytical tools jointly organised by the Bank of England, the European Central Bank, and the Data Analytics for Finance and Macro Research Centre at King's College London (London, UK)
April 2023: I'm presenting my working paper on NLP benchmarks for economics and finance at the Centre for Economic Research Policy (CEPR) Webinar Series
April 2023: Our working paper on multimodal NLP frameworks for central bank communication analysis got accepted to be presented at the Qatar Centre for Global Banking and Finance Annual Conference 2023
Feb 2023: I'm presenting a working paper on an NLP benchmark suite for economics and finance at the Economics Seminar at the University of Oxford (Oxford, UK)
Jan 2023: I'm honoured to have won an Alan Turing Institute research grant to work on "narrative analysis with multimodal language and graph models"
Selected Research
Ahrens, McMahon:
EcoFinBench - An NLP benchmark evaluation for economics and finance
working paper 2023
Ahrens, Erdemlioglu, McMahon, Neely, Yang:
Mind your language: market responses to central bank speeches
SSRN working paper 2023 (in submission to journal)
Gorduza, Ahrens, McMahon, Pierrehumbert:
Rumour detection with narrative graphs: using narrative-infused symbolic graphs to improve rumour detection in social media
working paper 2023
George, Ahrens, McMahon, Pierrehumbert:
Conspiracy detection with emotion markers: leveraging multidimensional content information to improve conspiracy detection
working paper 2023
Mathai , Ahrens, McMahon, Pierrehumbert:
NarrativeGraph: modelling text as a graph of micro-narratives
working paper 2023
Drinkall , Ahrens, McMahon, Pierrehumbert:
Narrative-informed information distillation of the Federal Open Market Committee meetings
working paper 2023
Ahrens, McMahon:
Extracting economic signals from central bank speeches
EMNLP 2021, ECONLP
Quian, Saunders, Ahrens:
Lawyering when the law becomes machine-learnt: mapping legaltech adoption and skill demand
book chapter in The Legaltech Book 2020
Ahrens:
Identifying monetary policy shocks with natural language processing
MPhil thesis 2018