ML Newsnew | past | best | rss | submit about
Stories from October 16, 2025 (UTC)
Go back a day, month, or year. Go forward a day.
1. TaxCalcBench: Evaluating Frontier Models on the Tax Calculation Task (arxiv.org)
70 points by handfuloflight 353 days ago | hide | past | pdf | 24 comments
2. LLMs struggle with math reasoning, because they can't conjecture (arxiv.org)
5 points by trehcrob 353 days ago | hide | past | pdf | 3 comments
3. LLMs Reproduce Human Purchase Intent via Semantic Similarity of Likert Ratings (arxiv.org)
4 points by sebg 353 days ago | hide | past | pdf | discuss
4. Tencent's Training-Free Group Relative Policy Optimization (arxiv.org)
3 points by felineflock 353 days ago | hide | past | pdf | discuss
5. Tensor Logic: The Language of AI (arxiv.org)
3 points by Anon84 353 days ago | hide | past | pdf | discuss
6. Towards Logic: The Language of AI (arxiv.org)
3 points by cmogni1 353 days ago | hide | past | pdf | discuss
7. The Art of Scaling Reinforcement Learning Compute for LLMs (arxiv.org)
2 points by sonabinu 353 days ago | hide | past | pdf | discuss
8. A Survey of Vibe Coding with Large Language Models (arxiv.org)
1 point by Gigacore 353 days ago | hide | past | pdf | discuss
About | RSS | RSS (all) | HN arXiv