 |
ML Newsnew | past | best | rss | submit | about |
|
Stories from February 8, 2024 (UTC)
|
| 1. | Grandmaster-Level Chess Without Search (arxiv.org) |
| 198 points by jonbaer on Feb 8, 2024 | hide | past | pdf | 130 comments | |
| 2. | Direct Language Model Alignment from Online AI Feedback (arxiv.org) |
| 61 points by drcwpl on Feb 8, 2024 | hide | past | pdf | 4 comments | |
| 3. | TransTroj: Transferable Backdoor Attacks to Pre-Trained Models (arxiv.org) |
| 22 points by indus on Feb 8, 2024 | hide | past | pdf | discuss | |
| 4. | Dot-product attention learns positional and semantic attention (arxiv.org) |
| 3 points by fzliu on Feb 8, 2024 | hide | past | pdf | discuss | |
| 5. | Long Is More for Alignment: A Simple Baseline for Instruction Fine-Tuning (arxiv.org) |
| 2 points by max-andr on Feb 8, 2024 | hide | past | pdf | 1 comment | |
| 6. | ReGAL: Refactoring Programs to Discover Generalizable Abstractions (arxiv.org) |
| 2 points by PaulHoule on Feb 8, 2024 | hide | past | pdf | discuss | |
| 7. | The Case for Co-Designing Model Architectures with Hardware (arxiv.org) |
| 1 point by PaulHoule on Feb 8, 2024 | hide | past | pdf | discuss | |