Publications
From meaning
to evidence.
Papers and preprints on multilingual semantics, semantic parsing, knowledge graphs, and predicting causal effects from research evidence. Open any paper to read its original abstract next to its first page.
- Papers & preprints
- 12
- Peer-reviewed
- 10
- Research threads
- 5
- Venues
- ACLNAACLLREC-COLINGAAAIVLDBDCMI
Research map
Five threads, one question: what does text mean?
Every paper, placed by year and coloured by theme. Pick a thread to follow it, or open any paper.
Selected
Start here
Three recommended first reads, one per thread. Every other paper and interactive story is in the map above and the full list below.
Paper by paper
All publications
2026
- DCMI 2026 · PostersposterEvidence & causal AI
Linxi Wang, Madeline Bassetti, Abelardo Carlos Martínez Lorenzo, Philipp Zimmer, Satvik Garg, Piriyakorn Piriyatamwong, Giuliano Martinelli, Saqib Hussain, Nihaa Sajid, Clémence Gall, Rida Khan, Aarushi Aggarwal, Jackson Mrema, Grace Sekwao, Jay Prakash, Dhruti Sanghavi, Shadrack Bentil, Mohammadmehdi Sharifkazemi, Sophie Pascale, Flavia Polles, Samuel Fraiberger
Presents ImpactAI, the World Bank’s generative AI evidence platform: a metadata layer over curated impact evaluations and a five-stage workflow (query understanding, retrieval, standardization, grounded generation, attribution) that returns source-grounded answers.
- arXiv preprintpreprintEvidence & causal AI
Giuliano Martinelli, Piriyakorn Piriyatamwong, Abelardo Carlos Martínez Lorenzo, Jasmin Baier, Riccardo Orlando, Satvik Garg, Sharif Kazemi, Linxi Wang, Arianna Legovini, Samuel P. Fraiberger
Introduces Query2Effect, a benchmark of more than 72,000 synthetic natural-language queries, and a two-step framework that structures queries before predicting causal effect sizes.
2025
- VLDB 2025 Workshops · LS-NSLworkshop paperKnowledge graphs
Abelardo Carlos Martínez Lorenzo, Alexander Perfilyev, Volker Markl, Martha Clokie, Thomas Sicheritz-Ponten, Zoi Kaoudi
Presents Poderoso, a modular framework combining knowledge-graph embeddings with rule-based reasoning for graph completion. The paper notes that the underlying work began during the author’s TU Berlin master’s thesis.
2024
- Findings of ACL 2024Multilingual resources
Abelardo Carlos Martínez Lorenzo, Pere-Lluís Huguet Cabot, Karim Ghonim, Lu Xu, Hee-Soo Choi, Alberte Fernández Castro, Roberto Navigli
Introduces the Multilingual Semantic Layer and a semi-automatic multilingual dataset creation pipeline, then shows that manually refining 1,100 sentences across 11 languages helps reduce the semantic parsing performance gap for low-resource languages.
- NAACL 2024Multilingual resources
Simone Conia, Edoardo Barba, Abelardo Carlos Martínez Lorenzo, Pere-Lluís Huguet Cabot, Riccardo Orlando, Luigi Procopio, Roberto Navigli
Presents a large multilingual corpus with hundreds of millions of silver semantic annotations across four NLU tasks and five languages, aiming to make explicit semantic supervision more available across languages and tasks.
- LREC-COLING 2024Semantic parsing
Abelardo Carlos Martínez Lorenzo, Roberto Navigli
Proposes a compact AMR linearization and a modular parser that cuts token counts by roughly 40 to 50 percent and reduces training and inference time by about 80 percent while preserving strong parsing quality.
2023
- Findings of ACL 2023Semantic parsing
Abelardo Carlos Martínez Lorenzo, Pere-Lluís Huguet Cabot, Roberto Navigli
Builds a cross-lingual AMR aligner that recovers span-to-graph alignments directly from transformer cross-attention, avoiding English-specific rules and improving alignment quality across multiple languages.
- ACL 2023Semantic parsing
Abelardo Carlos Martínez Lorenzo, Pere-Lluís Huguet Cabot, Roberto Navigli
Examines the weaknesses of existing AMR ensemble methods, adds validation for structural constraints, and proposes two transformer-based ensemble strategies that are both more robust and less computationally expensive.
- Findings of ACL 2023Semantic parsing
Pavlo Vasylenko, Pere-Lluís Huguet Cabot, Abelardo Carlos Martínez Lorenzo, Roberto Navigli
Introduces LeakDistill, which injects graph structure into transformer representations through structural adapters and word-to-node alignment, leading to state-of-the-art AMR parsing without extra training data.
2022
- ACL 2022Meaning representation
Abelardo Carlos Martínez Lorenzo, Marco Maru, Roberto Navigli
Introduces the BabelNet Meaning Representation as an interlingual formalism built on BabelNet and VerbAtlas, releases the BMR 1.0 dataset, and shows that the formalism supports strong multilingual parsing and generation.
- AAAI 2022Meaning representation
Roberto Navigli, Rexhina Blloshmi, Abelardo Carlos Martínez Lorenzo
Presents BMR as a broad, fully semantic and language-independent meaning representation designed to connect text, images, audio, video, and logical forms across languages and AI settings.
Doctoral research
Thesis
Sapienza University of Rome · 2024
Enhancing Semantic Parsing in the Age of Pre-trained Language Models
Doctoral thesis on multilingual meaning representation and efficient semantic parsing. The university record currently has no attached PDF.
University recordOther research output
Conference posters
DCMI 2026 · Poster and proceedings
ImpactAI: Embedding Generative AI into Evidence-Informed Development Decision-Making
Co-authored poster on ImpactAI’s evidence-grounded design, use cases, and deployment experience.
UniDive 2nd General Meeting · 2024
Creating a Multilingual Wide-Coverage Multi-Layered Semantically Annotated Corpus
A poster related to the MOSAICo corpus, listed separately from the NAACL paper above.