Model excels at Sokoban-style puzzles, sparking questions about training data contamination
lukaszkaiser · x · 2026-09-26
Yoav Goldberg observes a frontier model is very good at grid-based, turn-based Sokoban-like puzzle games and asks whether it was trained on such games. Łukasz Kaiser guesses probably not specifically, but possibly some, noting that training with controlled data would be great for research.
Related event: Frontier Model Excels at Sokoban Puzzles, Raising Training Data Questions(2 posts)→
More from Models
- GPT 6 Luna Max one-shots porting an Nvidia project to wgpu shaders — and it's faster — mgostIH · 2026-09-26
- Observation: Astra uses filler tokens far more effectively than other models — scaling01 · 2026-09-26
- How Long Until Local ~30B A3B Models Match GLM 5.3 Flash Quality? — Aggravating-Push-207 · 2026-09-26
- Ethan Mollick: 'Keep prompts short' is bad advice, and minimizing token cost confuses inputs with outputs — emollick · 2026-09-26
- Matthew Berman Reviews DeepSeek and Calls It 'Crazy' — Matthew Berman · 2026-09-26
- GPT-6 Luna uses fewer reasoning tokens than 5.6 on ARC-AGI-2, hard tasks stymie both — mhmazur · 2026-09-26