← Back to People
A machine learning researcher working on scalable data methods for language models, from pre-training to post-training.
3
Mentions
3
Articles
3
Outlets
Feb 2026 — Oct 2026
Coverage
Positive
Sentiment
Associated Topics
Companies Covered
Coverage Patterns
How media typically covers Luca Soldaini
Article Types
- research1
Media Angles
- technical1
Narrative Framing
- breakthrough1
Appears In
Sentiment Distribution
Based on 3 scored articles
Positive 100%
Associated AI Models
ChatGPT1Frontier models (distilled to 8B open model)1
Appears Alongside
Articles
Most recent firstMentioned In
Referenced in coverage
Byteification retrofits subword LLMs to read raw bytes
NaturepositiveOct 8, 2026
“Co-authored Nature paper on byte-level language models and model conversion methods.”
Allen AI releases open-source toolkit to convert messy PDFs into clean LLM training data
GitHubpositiveJul 2, 2026
“Developer and contributor to olmOCR project.”
How2Everything: Mining the Web for How-To Procedures to Evaluate and Improve LLMs
Allen Institute for AI (Ai2)researchpositiveFeb 15, 2026
How2Everything, a new framework for mining 351K how-to procedures from the web, enables scalable evaluation and improvement of LLM procedural generation through an LLM judge that achieves 80.5% human agreement and yields measurable performance gains via reinforcement learning.
“Co-authored How2Everything framework for evaluating and improving LLM procedure generation.”
LLMsAI ResearchFoundation ModelsReinforcement Learning