|
|
|
Stories from May 15, 2025 (UTC)
|
| 1. |
LLMs get lost in multi-turn conversation (arxiv.org) |
|
374 points by simonpure on May 15, 2025 | hide | past | pdf | 259 comments
|
|
| 2. |
Self Rewarding Self Improving: Autonomous LLM Improvement (arxiv.org) |
|
28 points by tamassimond on May 15, 2025 | hide | past | pdf | discuss
|
|
| 3. |
DeepSeek-V3: Achieving Efficient LLM Scaling with 2,048 GPUs (arxiv.org) |
|
7 points by qtwhat on May 15, 2025 | hide | past | pdf | 1 comment
|
|
| 4. |
Why do LLMs attend to the first token? (arxiv.org) |
|
2 points by adhi01 on May 15, 2025 | hide | past | pdf | 1 comment
|
|
| 5. |
A General Theoretical Paradigm to Understand Learning from Human Preferences (arxiv.org) |
|
2 points by yenniejun111 on May 15, 2025 | hide | past | pdf | discuss
|
|
| 6. |
Online Isolation Forest (arxiv.org) |
|
2 points by badmonster on May 15, 2025 | hide | past | pdf | discuss
|
|
| 7. |
Understanding Perception and Reasoning Through Model Merging (arxiv.org) |
|
2 points by veryluckyxyz on May 15, 2025 | hide | past | pdf | discuss
|
|
| 8. |
Language Agents Mirror Human Causal Reasoning Biases. How Can We Help Them Think (arxiv.org) |
|
1 point by badmonster on May 15, 2025 | hide | past | pdf | discuss
|
|
| 9. |
Insights into DeepSeek-V3: Scaling Challenges and Reflections on Hardware for AI (arxiv.org) |
|
1 point by matt_d on May 15, 2025 | hide | past | pdf | discuss
|
|
|
|
