Hello! I am a machine learning engineer at Microsoft’s Office for Applied Research.

Research Interests:

• Uncertainty Quantification & Active Learning with Professor Jackie Cheung at MILA and McGill University (MSc, Research Track).
• Summarization and Simplification with Prof. Arman Cohan and Prof. Dragomir Radev, Linyong Nan at Yale University (BSc)

Past Work Experience:

• Internships: Adobe (2025, AI Applied Research), Elicit (2024, ML Engg)
• Work: McKinsey and Company, QuantumBlack (2023-2024, Data Science)

Selected Publications

• Lorenzo Flores, Cesare Spinoso di-Piano, Ori Ernst, David Ifeoluwa Adelani, and Jackie Chi Kit Cheung. Testing the Assumptions of Active Learning for Translation Tasks with Few Samples, EMNLP 2026 Multilingual Representation Learning Workshop [Paper, Code]
• Lorenzo Flores, Cesare Spinoso di-Piano, Jackie Chi Kit Cheung. Confident in a Confidence Score: Investigating the Sensitivity of Confidence Scores to Supervised Fine-Tuning, EMNLP 2026 UncertaiNLP Workshop [Paper, Code]
• Lorenzo Flores, Junyi Shen, Goodman Gu. Towards Reliable Multi-Agent Systems for Marketing Applications via Reflection, Memory, and Planning, ArXiv [Paper]
• Lorenzo Flores, Ori Ernst, Jackie Chi Kit Cheung. Improving the Calibration of Confidence Scores in Text Generation Using the Output Distribution’s Characteristics, ACL 2025 [Paper, Code]
• Lorenzo Flores and Arman Cohan. On the Benefits of Fine-Grained Loss Truncation: A Case Study on Factuality in Summarization, EACL 2024 [Paper, Video, Code]
• Lorenzo Flores, Heyuan Huang, Kejian Shi, Sophie Chheang, and Arman Cohan. 2023. Medical Text Simplification: Optimizing for Readability with Unlikelihood Training and Reranked Beam Search Decoding, EMNLP 2023 Findings [Paper, Video, Code, Demo]
• Linyong Nan, Lorenzo Flores, Yilun Zhao, Yixin Liu, Luke Benson, Weijin Zou, and Dragomir Radev. 2022. R2D2: Robust Data-to-Text with Replacement Detection, EMNLP 2022 [Paper]
• Lorenzo Flores, Dragomir Radev. 2022. Look Ma, Only 400 Samples! Revisiting the Effectiveness of Automatic N-Gram Rule Generation for Spelling Normalization in Filipino, EMNLP 2022 SustaiNLP Workshop [Paper, Video, Code]
• Lorenzo Flores, Yiding Hao. 2022. Adversarial Benchmark for Fake News Classification. AAAI 2022 AdvML Workshop [Paper, Code]
• Chiara Ledesma, Oshean Lee Garonita, Lorenzo Flores, Isabelle Tingzon, and Danielle Dalisay. 2020. Interpretable Poverty Mapping using Social Media Data, Satellite Images, and Geospatial Information, ML4D Workshop, NeurIPS 2020, Best Workshop Paper Award [Paper]

Projects

• WriteDoc (check us out at write-doc.com!): a medical scribe + paperwork tool for Filipino/Taglish! We’re piloting with various doctors + clinics – would love to chat if you’re interested in healthcare x AI [Here!]
• LossLibrary: a repository that consolidates loss functions from NLP literature and helps users integrate it into training [Here!]