Tag
Researchers at Binghamton University used Shannon entropy to develop a mathematical method that solves Wordle puzzles with a 99% success rate, prioritizing informative guesses over likely answers.
This paper introduces RecurrReason, a difficulty-controlled benchmark of four symbolic logic puzzles to evaluate multi-step reasoning in sequence models. Fine-tuning experiments on T5 and GPT-2 show that architecture determines success more than scale, and that pre-training transfer depends on local transition structure.