| 1. |
Extracting memorized pieces of books from open-weight language models (arxiv.org) |
|
109 points by fzliu on Jun 16, 2025 | hide | past | pdf | 109 comments
|
|
| 2. |
Breaking Quadratic Barriers: A Non-Attention LLM for Ultra-Long Context Horizons (arxiv.org) |
|
70 points by PaulHoule on Jun 16, 2025 | hide | past | pdf | 23 comments
|
|
| 3. |
Spherical CNNs (2018) (arxiv.org) |
|
22 points by rkp8000 on Jun 16, 2025 | hide | past | pdf | 3 comments
|
|
| 4. |
The Illusion of the Illusion of Thinking – A Comment on Shojaee et al. (2025) (arxiv.org) |
|
16 points by gfortaine on Jun 16, 2025 | hide | past | pdf | 14 comments
|
|
| 5. |
The Illusion of the Illusion of Thinking (arxiv.org) |
|
12 points by jedisct1 on Jun 16, 2025 | hide | past | pdf | 1 comment
|
|
| 6. |
Towards Understanding Sycophancy in Language Models (arxiv.org) |
|
9 points by fzliu on Jun 16, 2025 | hide | past | pdf | 2 comments
|
|
| 7. |
Vision Transformers Don't Need Trained Registers (arxiv.org) |
|
4 points by avd4292 on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 8. |
How Do Olympiad Medalists Judge LLMs in Competitive Programming? (arxiv.org) |
|
3 points by npalli on Jun 16, 2025 | hide | past | pdf | 1 comment
|
|
| 9. |
BanglaByT5: Byte-Level Modelling for Bangla (arxiv.org) |
|
2 points by PaulHoule on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 10. |
Improving Continual Pre-Training Through Seamless Data Packing (arxiv.org) |
|
2 points by PaulHoule on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 11. |
Lossless Token Sequence Compression via Meta-Tokens (arxiv.org) |
|
2 points by PaulHoule on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 12. |
LiveCodeBench Pro: How Olympiad Medalists Judge LLMs in Competitive Programming? (arxiv.org) |
|
2 points by EvgeniyZh on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 13. |
Improving Brain-to-Image Reconstruction via Fine-Grained Text Bridging (arxiv.org) |
|
1 point by PaulHoule on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 14. |
Revealing Political Bias in LLMs Through Structured Multi-Agent Debate (arxiv.org) |
|
1 point by rntn on Jun 16, 2025 | hide | past | pdf | discuss
|
|
| 15. |
Appraisal-Based Chain-of-Emotion Improves AI Persona Accuracy (arxiv.org) |
|
1 point by virtual_rf on Jun 16, 2025 | hide | past | pdf | discuss
|
|
