LLMs 10
- Harness improvements help deploy a model better, but the model's underlying intelligence remains the core driver of capability.
- The deployment layer surrounding a model is as important as the model's raw intelligence.
Lilian Weng
Machine-learning researcher; has written the Lil'Log survey posts on how a model technique works since 2017, and worked at OpenAI from 2018 to 2024.
Lilian Weng did not write this page.
We collected these quotes from things they published elsewhere, and every quote links to where it was said. They have no account here and have not endorsed this site. Quotes are word for word; the short line under each one is our own restatement, not their wording. Their own site. Is this you? Claim it or ask us to remove it. Or tell us what is wrong here.
What Lilian Weng says most about
cybersecurity 3
- Longer test-time thinking improves an AI model's robustness to adversarial or unusual inputs.
- Higher-perplexity adversarial attack text is easier to detect and mitigate.
AI alignment 2
- More capable AI agents are more likely to find and exploit flaws in their reward functions.
- It is fundamentally difficult to design a reward function that accurately captures the intended goal in reinforcement learning.
design 2
- More capable AI agents are more likely to find and exploit flaws in their reward functions.
- It is fundamentally difficult to design a reward function that accurately captures the intended goal in reinforcement learning.
neural networks 2
- Longer test-time thinking improves an AI model's robustness to adversarial or unusual inputs.
- As models keep growing, AI is running out of enough high-quality unique training tokens to keep up.
reinforcement learning 2
- More capable AI agents are more likely to find and exploit flaws in their reward functions.
- It is fundamentally difficult to design a reward function that accurately captures the intended goal in reinforcement learning.