Reasoning models' Towers of Hanoi failures persist under stepwise prompting, while River Crossing failures mostly vanish when tests are restricted to solvable configurations.
For each configuration, we run 10 independent trials to obtain robust averages
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Rethinking the Illusion of Thinking
Reasoning models' Towers of Hanoi failures persist under stepwise prompting, while River Crossing failures mostly vanish when tests are restricted to solvable configurations.