A The ARC Atlas

Appendix A · Canonical bibliography

Every Work, listed once.

This is the book's canonical bibliography of papers, reports, posts, and code releases. Each Work keeps one title, public credit line, year, venue, type, and primary URL; the Atlas adds its editorial reading separately.

25
Works
24
With Atlas entries
1
Supplemental Work
4
Publication types

Work index

One bibliography, no duplicate records.

A catalog number appears only when a Work also has an Atlas entry. The supplemental ARC-AGI-3 paper still belongs in the bibliography without being forced into the 21-entry editorial catalog.

Work Year / type Public credit line Atlas Primary source
p01 On the Measure of Intelligence arXiv · 2019 2019Papers François Chollet (Google) №01 → Open ↗
p02 ARC-AGI-2: A New Challenge for Frontier AI Reasoning Systems arXiv · 2025 2025Papers François Chollet, Mike Knoop, Gregory Kamradt, Bryan Landers, Henry Pinkard (ARC Prize Foundation) №02 → Open ↗
p03 ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence arXiv · 2026 2026Papers ARC Prize Foundation (benchmark design by François Chollet; co-founders Mike Knoop, Gregory Kamradt; lead environment design Hunter Henry; with the ARC-AGI-3 game-studio team) №03 → Open ↗
p04 ARC Prize 2024: Technical Report arXiv · 2024 2024Reports François Chollet, Mike Knoop, Gregory Kamradt, Bryan Landers (ARC Prize Foundation) №04 → Open ↗
p05 ARC Prize 2025: Technical Report arXiv · 2026 2026Reports ARC Prize Foundation (François Chollet, Mike Knoop, Gregory Kamradt, Bryan Landers) №05 → Open ↗
p06 OpenAI o3 Breakthrough High Score on ARC-AGI-Pub ARC Prize blog · 2024 2024Blog posts ARC Prize (François Chollet) №06 → Open ↗
p07 ARC 2020 Kaggle 1st-Place Solution (Icecuber) Kaggle ARC · 2020 2020Code Johan S. Wind ("icecuber") №07 → Open ↗
p08 The Surprising Effectiveness of Test-Time Training for Abstract Reasoning arXiv · 2024 · ICML 2025 2024Papers Ekin Akyürek, Mehul Damani, Adam Zweiger, Linlu Qiu, Han Guo, Jyothish Pari, Yoon Kim, Jacob Andreas (MIT CSAIL) №08 → Open ↗
p09 Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective arXiv · ICML 2025 2025Papers Daniel Franzen, Jan Disselhoff, David Hartmann ("the ARChitects") №09 → Open ↗
p10 Getting 50% (SoTA) on ARC-AGI with GPT-4o Redwood Research blog · 2024 2024Blog posts Ryan Greenblatt (Redwood Research) №10 → Open ↗
p11 How I got a record 53.6% on ARC-AGI-Pub using Sonnet 3.5 with Evolutionary Test-time Compute Substack · 2024 2024Blog posts Jeremy Berman №11 → Open ↗
p12 Combining Induction and Transduction for Abstract Reasoning ICLR 2025 2024Papers Wen-Ding Li, Keya Hu, Carter Larsen, Yuqing Wu, Simon Alford, Caleb Woo, Spencer M. Dunn, Hao Tang, Michelangelo Naim, Dat Nguyen, Wei-Long Zheng, Zenna Tavares, Yewen Pu, Kevin Ellis (Cornell, Shanghai Jiao Tong University, Basis, Autodesk) №12 → Open ↗
p13 Searching Latent Program Spaces (LPN) arXiv 2024 · NeurIPS 2025 spotlight 2024Papers Clément Bonnet, Matthew V. Macfarlane №13 → Open ↗
p14 Hypothesis Search: Inductive Reasoning with Language Models ICLR 2024 2023Papers Ruocheng Wang, Eric Zelikman, Gabriel Poesia, Yewen Pu, Nick Haber, Noah D. Goodman (Stanford University; Yewen Pu at Autodesk Research). Wang and Zelikman contributed equally. №14 → Open ↗
p15 CodeIt: Self-Improving Language Models with Prioritized Hindsight Replay ICML 2024 2024Papers Natasha Butt, Blazej Manczak, Auke Wiggers, Corrado Rainone, David W. Zhang, Michaël Defferrard, Taco Cohen (Qualcomm AI Research; some authors also University of Amsterdam). №15 → Open ↗
p16 ARC-AGI Without Pretraining (CompressARC) arXiv · 2025 2025Papers Isaac Liao, Albert Gu (Carnegie Mellon University). №16 → Open ↗
p17 Graph-Based Exploration for ARC-AGI-3 Interactive Reasoning Tasks arXiv · 2025 2025Papers Evgenii Rudakov, Jonathan Shock, Benjamin Ultan Cowley №17 → Open ↗
p18 DreamCoder: Growing generalizable, interpretable knowledge with wake-sleep Bayesian program learning arXiv · 2020 2020Papers Kevin Ellis, Catherine Wong, Maxwell Nye, Mathias Sablé-Meyer, Luc Cary, Lucas Morales, Luke Hewitt, Armando Solar-Lezama, Joshua B. Tenenbaum №18 → Open ↗
p19 Large Language Models as General Pattern Machines CoRL 2023 2023Papers Suvir Mirchandani, Fei Xia, Pete Florence, Brian Ichter, Danny Driess, Montserrat Gonzalez Arenas, Kanishka Rao, Dorsa Sadigh, Andy Zeng (Google DeepMind / Stanford) №19 → Open ↗
p20 Understanding and Benchmarking Artificial Intelligence: OpenAI's o3 Is Not AGI arXiv · 2025 2025Papers Rolf Pfister, Hansueli Jud №23 → Open ↗
p21 Don't throw the baby out with the bathwater: How and why deep learning for ARC arXiv · 2025 2025Papers Jack Cole, Mohamed Osman №24 → Open ↗
hill-climbing-arc-agi-3 Hill-Climbing ARC-AGI-3 ARC-AGI-3 academic partner paper · 2026 2026Papers Alexis Fox, Junlin Wang, Paul Rosu, Bhuwan Dhingra Supplemental No primary URL
p22 Executable World Models for ARC-AGI-3 in the Era of Coding Agents arXiv · AGI-2026 2026Papers Sergey Rodionov №20 → Open ↗
p23 OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3 arXiv · 2026 2026Papers David Courtis, Wenhao Li, Scott Sanner №21 → Open ↗
p24 Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3? arXiv · 2026 2026Papers Sergey Rodionov №22 → Open ↗

YouTube first · local synths follow

ARC Radio

01 / 12 🦉 8-Bit Chiptune Playlist 🦉 Retro Video Game Music for Nostalgic Vibes YouTube · external stream

The 4 requested YouTube selections play first and require a network connection; their titles refresh from YouTube when they load. 8 original AI-composed retro-game loops follow and are generated live in your browser. Audio keeps playing when you close this panel and stops only when you press Pause.

Field notes · reader review

Help improve this guide

Found a wrong score, broken link, missing paper, or unclear passage? Tell us what you noticed.

How useful is it? optional
- / 5
What kind of note? optional

No account, no tracking. Sent straight to the maintainer.