ICLR 2026 Affiliation Dataset: PDF-derived institutional data for 5,356 accepted papers with treemap visualizations
By
stared
Kettled twice. Extra chewy, extra trustworthy.
Summary
A GitHub repository provides an end-to-end pipeline that extracts institutional affiliations from the PDF title blocks of 5,356 ICLR 2026 accepted papers, avoiding the OpenReview profile drift problem. The project delivers a clean dataset (CSV + XLSX) and treemap visualizations showing which institutions are shaping AI research, with affiliations sourced directly from paper PDFs rather than author profiles.
Key quotes
· 3 pulledThis avoids the OpenReview-profile drift problem (where authors' current job appears on every paper they ever wrote — e.g. listing Wyoming as the affiliation for a paper actually written at UBC).
Affiliations come from the paper's title block PDF, not from author profiles.
End-to-end pipeline that turns 5,356 ICLR 2026 accepted papers into a clean, PDF-derived institutional-affiliation dataset and a publication-ready treemap of who is shaping AI research right now.
You might also wanna read
Talkie: A 13B Vintage Language Model Trained on 1930s Texts
This article introduces "talkie-1930-13b-it," a vintage language model trained on texts from the 1930s, designed to simulate conversation wi
talkie-lm.com·1mo ago
Introduction to Machine Learning: Visual Guide to Classification with Home Data Example
This article provides an introductory, visual explanation of machine learning concepts using a practical example of classifying homes in New
r2d3.us·2mo ago
Introduction to Decision Trees: Understanding Entropy and Information Gain in Machine Learning
This article provides an introduction to decision trees, focusing on entropy and information gain concepts in machine learning. It explains
mlu-explain.github.io·3mo ago
Anthropic's AI Fluency Index: Measuring How People Develop AI Collaboration Skills
Anthropic's AI Fluency Index report examines how people develop AI collaboration skills through observable behaviors in Claude.ai conversati
anthropic.com·3mo ago
Measuring Data Processing Effectiveness: Defining Insight and Compression Efficiency
The article discusses methods for measuring how much data a person can effectively process or understand, focusing on defining 'insight' as
news.ycombinator.com·3mo ago
Research on AI Assistance and Coding Skill Development: Productivity vs. Learning Trade-offs
The article discusses research on how AI assistance impacts coding skill development, examining whether AI tools that increase productivity
anthropic.com·4mo ago