Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks — mentioned by puttycat

Tue, 08 Jul 2025 15:03:38 GMT

If LLMs can't even do basic arithmetic [1,2,3], why should anyone trust them with what they put in their mouths? [1] https://arxiv.org/abs/2410.21272 [2] https://arxiv.org/abs/2406.02356 [3] https://github.com/0xnurl/gpts-cant-count

Paper and discussion history. A mention is not an endorsement.

Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks · HackerPapers discussions

Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks — mentioned by puttycat