Evidence synthesis
Building systems that can retrieve, structure, and summarize rigorous findings from development research.
Lead AI Scientist
Abelardo Carlos Martinez Lorenzo
AI for Good (Development Economics)
Lead AI Scientist at the World Bank building evidence synthesis systems, scientific information extraction pipelines, and domain-specialized language models for development policy.
Through ImpactAI, I work on systems that extract, structure, and reason over development research so policymakers can work with rigorous evidence from randomized controlled trials more effectively.
My background spans multilingual NLP, semantic parsing, scientific information extraction, and applied ML infrastructure, with research presented at ACL, NAACL, COLING, and AAAI.
Current focus
Research systems for evidence synthesis and policy intelligence at the World Bank.
AI leadership
Leading ImpactAI work on evidence synthesis and policy intelligence at the World Bank.
Publication record
ACL, NAACL, COLING, AAAI, multilingual NLP, semantic parsing, and scientific information extraction.
Research training
Marie Skłodowska-Curie Fellow with a Ph.D. in NLP and multilingual semantic technologies.
Research profile
From multilingual NLP to evidence systems
My work sits between research and delivery: building systems that retrieve, interpret, and synthesize scientific evidence while remaining useful for real decision-making.
I care about domain-specialized language models, structured knowledge extraction, evidence synthesis, and AI systems that make public policy more rigorous and more accessible.
Domain-specialized LLMs
Adapting large language models to scientific and policy contexts where precision and traceability matter.
AI for policy
Translating machine learning research into tools that improve how institutions reason and decide.
About
Research, engineering, and policy applications in one track
Abelardo's work connects multilingual NLP and semantic parsing research with production systems for evidence synthesis. The same thread runs from Sapienza and Babelscape to the World Bank: make complex scientific knowledge easier to retrieve, structure, and use in practice.
Current leadership
Lead AI Scientist at the World Bank, building systems for evidence synthesis and policy intelligence through ImpactAI.
Research training
Ph.D. in Deep Learning and Natural Language Processing from Sapienza University of Rome as a Marie Skłodowska-Curie Fellow.
Research DNA
Multilingual NLP, semantic parsing, scientific information extraction, and domain-specialized language models.
Career path
Lead AI Scientist (Team Lead + Senior Research Engineer)
The World Bank · ImpactAI
Sept 2024 – Present · Washington, DC, USA
Lead a six-person AI science team delivering an agentic framework that synthesizes evidence from RCTs into policy-grade meta-analyses.
Senior AI Scientist
The World Bank
Jan 2024 – Sept 2024 · Washington, DC, USA
Built domain-specific entity linking models (bi-encoder DeBERTa + e5) to map text into impact-evaluation taxonomies.
NLP Researcher
Sapienza NLP · Sapienza University
2023 – 2024 · Rome, Italy
Led semantic parsing research and mentored MSc students on multilingual NLU projects.
NLP Researcher
Babelscape
2020 – 2023 · Rome, Italy
Designed cross-lingual semantic representation frameworks and state-of-the-art AMR parsing ensembles.
Education
Ph.D. in Deep Learning & Natural Language Processing (Marie Skłodowska-Curie Fellow)
Sapienza University of Rome
Rome, Italy
Project: EU-Horizon 2020 KnowGraphs · Thesis: Enhancing Semantic Parsing in the Age of Pre-trained Language Models.
MSc in Big Data Management and Analytics (BDMA)
Erasmus Mundus Joint Master Programme
Berlin · Barcelona · Brussels
Final Grade: 9.0/10.
BSc in Software Engineering
Universidad Málaga
Málaga, Spain
Final Grade: 9.0/10 · Erasmus Exchange: Univerzita Karlova, Prague.
Outside the lab
Photography, travel, and visual curiosity
Photography
A visual practice shaped by light, streets, landscapes, and the details that tell a story.
Travel
I like moving across cities, coasts, and mountain routes to learn through people, food, and place.
Art, culture, and history
Museums, architecture, and historical spaces sharpen how I think about context, memory, and design.