ML Newsnew | past | best | rss | submit about
Stories from September 16, 2026 (UTC)
Go back a day, month, or year. Go forward a day.
1. Breaking the 1.58-bit Barrier for Ternary LLMs (arxiv.org)
245 points by matt_d 18 days ago | hide | past | pdf | 41 comments
2. Dream-RSI: Recursive Self-Improvement through Evolving Worlds (arxiv.org)
213 points by bananaflag 18 days ago | hide | past | pdf | 53 comments
3. How good are frontier models at physics? (arxiv.org)
100 points by qt31415926 18 days ago | hide | past | pdf | 50 comments
4. Accurate Models of AMD Matrix Cores (arxiv.org)
80 points by matt_d 18 days ago | hide | past | pdf | 11 comments
5. Potemkin Understanding in Large Language Models (2025) (arxiv.org)
3 points by mpweiher 18 days ago | hide | past | pdf | 1 comment
6. Competitive Market Behavior of LLMs (arxiv.org)
2 points by paulpauper 18 days ago | hide | past | pdf | discuss
7. Coding Agents Have Converged: Why the SWE-Bench Leaderboard Can No Longer Order (arxiv.org)
2 points by sbulaev 18 days ago | hide | past | pdf | discuss
8. Continual Learning Mechanisms Compose for Long-Horizon Memorization (arxiv.org)
2 points by donk8r 18 days ago | hide | past | pdf | discuss
9. Divide, Consult, Conquer: Capability Laundering Through Aligned LLMs (arxiv.org)
2 points by sbulaev 19 days ago | hide | past | pdf | discuss
10. Fractal basins trap latent reasoning (arxiv.org)
1 point by wil3 18 days ago | hide | past | pdf | discuss
11. Occamy-1.0: Open Pareto-frontier 35B intelligence for co-work (arxiv.org)
1 point by donk8r 18 days ago | hide | past | pdf | discuss
About | RSS | RSS (all) | HN arXiv