Tag: utility
Counterfactuals against sycophancy
2026-08-17 CoT evaluation prompting text truthfulness utility Evaluating how chatbots confirm the user's feelings rather than the truth, but in a way that doesn't require the truth to actually exist. Access: $ Basic
Automatic prompt injection
2026-07-20 LLaMA applications prompting text tokenization training utility Discrete optimization can be used to create universal prompt injections that subvert multiple different task prompts. Access: $ Basic
Best of "Research Ideas"
2026-07-13 alignment evaluation meta prompting theory training utility I run down the 15 best of my Research Ideas suggestions for the past year. Access: $ Basic
Twenty qwestions from Qwen3.6
2026-07-06 Qwen applications interpretation meta utility I answer 20 questions suggested by Qwen3.6 run on the transcripts of the last year's talks, with commentary on what these questions imply about the model itself Access: Free account