 |
ML Newsnew | past | best | rss | submit | about |
|
Stories from March 3, 2025 (UTC)
|
| 1. | Cautious Optimizers: Improving Training with One Line of Code (arxiv.org) |
| 66 points by tosh on Mar 3, 2025 | hide | past | pdf | 2 comments | |
| 2. | Order Doesn’t Matter, But Reasoning Does (arxiv.org) |
| 14 points by spaintech on Mar 3, 2025 | hide | past | pdf | 16 comments | |
| 3. | Art: Anonymous Region Transformer (arxiv.org) |
| 2 points by geox on Mar 3, 2025 | hide | past | pdf | discuss | |
| 4. | Transformers Learn to Implement Multistep Gradient Descent with Chain of Thought (arxiv.org) |
| 1 point by bearseascape on Mar 3, 2025 | hide | past | pdf | discuss | |
| 5. | Flash Interpretability: Decoding Specialised Feature Neurons in LLM (arxiv.org) |
| 1 point by mococa on Mar 3, 2025 | hide | past | pdf | discuss | |
| 6. | Mixtera: A Data Plane for Foundation Model Training (arxiv.org) |
| 1 point by mboether on Mar 3, 2025 | hide | past | pdf | discuss | |