DeepMind researcher suggests LLMs can actively discover new abstractions

Google DeepMind researcher Andrew Lampinen argues that language models can show more intentional generalization, such as realizing cheat codes likely exist in games and actively searching for them, suggesting they may discover new abstractions like mathematicians.

2026-10-03 ~ 2026-10-03 · 2 related posts

• Researcher: LMs may 'realize' cheat codes exist and deliberately search for them — AndrewLampinen · 2026-10-03
• DeepMind's Lampinen: LMs may make 'intentional' discoveries of new math concepts — AndrewLampinen · 2026-10-03