Probing 2D spatial reasoning in text-only LLMs: state tracking beats fixed plans
Ashwin Nedungadi · hf · 2026-09-03
A study by Ashwin Nedungadi probes open-ended 2D layout generation in text-only LLMs.
- Models vary widely in open-ended 2D layout ability, even though geometry-to-code translation is reliable.
- Performance depends heavily on the output medium and on the model's internal tracking of an evolving geometric state.
- Better models maintain spatial state while generating rather than committing to a fixed plan upfront.
More from Research
- Eigenfaces revisited: interactive demos of the pioneering computer vision technique — CSProfKGD · 2026-09-03
- Three-Paper Series Decomposes Human-Like RSI into ASPIRE, S³Gym and HarnessDev — teortaxesTex · 2026-09-03
- Noematrix's Noe-0 trains embodied models with zero teleoperation data — jiqizhixin · 2026-09-03
- New research shows user feedback signals like "that's wrong" can be leveraged for training — LChoshen · 2026-09-03
- Study finds all 13 major LLMs flip truth judgments on speaker gender, up to 23.6% of statements — anthara_ai · 2026-09-03
- R³ paper: robots learn to think before acting with carefully done RL on explaining demo data — aviral_kumar2 · 2026-09-03