|
|
|
Stories from October 10, 2023 (UTC)
|
| 1. |
HyperAttention: Long-Context Attention in Near-Linear Time (arxiv.org) |
|
73 points by kelseyfrog on Oct 10, 2023 | hide | past | pdf | 13 comments
|
|
| 2. |
Outlier Weighed Layerwise Sparsity: A Missing Secret Sauce for Pruning LLMs (arxiv.org) |
|
5 points by amilios on Oct 10, 2023 | hide | past | pdf | 1 comment
|
|
| 3. |
Fast Neural Rendering with Multi-Input Multi-Output Neural Radiance Fields (arxiv.org) |
|
3 points by PaulHoule on Oct 10, 2023 | hide | past | pdf | discuss
|
|
| 4. |
Finetuning 3-Bit LLMs on Consumer GPUs by Integrating with Modular Quantizers (arxiv.org) |
|
2 points by volodia on Oct 10, 2023 | hide | past | pdf | discuss
|
|
| 5. |
Just Ask for Calibration: Eliciting Calibrated Confidence Scores from LLMs (arxiv.org) |
|
2 points by famouswaffles on Oct 10, 2023 | hide | past | pdf | discuss
|
|
| 6. |
Grokking as Compression: A Nonlinear Complexity Perspective (arxiv.org) |
|
2 points by amilios on Oct 10, 2023 | hide | past | pdf | 1 comment
|
|
| 7. |
Why do we need weight decay in modern deep learning? (arxiv.org) |
|
2 points by max-andr on Oct 10, 2023 | hide | past | pdf | 1 comment
|
|
| 8. |
Self-Taught Optimizer (Stop): Recursively Self-Improving Code Generation (arxiv.org) |
|
1 point by haltist on Oct 10, 2023 | hide | past | pdf | discuss
|
|
| 9. |
Enabling Level-4 Autonomous Driving on a Single $1k Off-the-Shelf Card (arxiv.org) |
|
1 point by robertlagrant on Oct 10, 2023 | hide | past | pdf | discuss
|
|
| 10. |
PB-LLM: Partially Binarized Large Language Models (arxiv.org) |
|
1 point by tosh on Oct 10, 2023 | hide | past | pdf | discuss
|
|
|
|
