|
|
|
Stories from September 8, 2025 (UTC)
|
| 1. |
Refrag: Rethinking RAG Based Decoding (arxiv.org) |
|
4 points by itchyjunk on Sep 8, 2025 | hide | past | pdf | discuss
|
|
| 2. |
KVComp: A High-Performance, LLM-Aware, Lossy Compression Framework for KV Cache (arxiv.org) |
|
3 points by kstonekuan on Sep 8, 2025 | hide | past | pdf | discuss
|
|
| 3. |
Why Language Models Hallucinate (arxiv.org) |
|
3 points by sonabinu on Sep 8, 2025 | hide | past | pdf | 1 comment
|
|
| 4. |
Geometric Deep Learning Grids, Groups, Graphs, Geodesics, and Gauges [pdf] (arxiv.org) |
|
3 points by ibobev on Sep 8, 2025 | hide | past | pdf | discuss
|
|
| 5. |
How to Hack Transformers: Steering LLMs via Prompts, States, and Weight Edits (arxiv.org) |
|
2 points by WASDAai on Sep 8, 2025 | hide | past | pdf | 1 comment
|
|
| 6. |
A Comprehensive Survey on Trustworthiness in Reasoning with LLMs (arxiv.org) |
|
1 point by Anon84 on Sep 8, 2025 | hide | past | pdf | discuss
|
|
| 7. |
The LLM Has Left the Chat: Evidence of Bail Preferences in LLMs (arxiv.org) |
|
1 point by geox on Sep 8, 2025 | hide | past | pdf | discuss
|
|
|
|
