That's called sleeper agent problem, and is extremely actual (and I don't think solvable): x.com/karpathy/status/17459212050207994…...

Andrej Karpathy's comment: twitter.com/karpathy/status/17459212050…

If you are using no-code solutions, increasing an "idea" in a dataset will make that idea more likely to appear. If you are fine-tuning your own LLM, there are other ways to get your idea to appear. In the literature this is sometimes called RLHF or preference optimization, and here are a few approaches: Direct Preference Optimization This uses Elo-scores to learn pairwise preferences. Elo is used in chess and basketball to rank individuals who compete in pairs. @argilla_io on X.com has been doing some work in evaluating DPO. Here is a decent thread on this: x.com/argilla_io/status/174505757169669… Identity Preference Optimization IPO is research from Google DeepMind. It removes the reliance of Elo scores to address overfitting issues in DPO. Paper: x.com/kylemarieb/status/172828158130623… Kahneman-Tversky Optimization KTO is an approach that uses mono preference data. For example, it asks if a response is "good or not." This is helpful for a lot of real word situations (e.g. "Is the restaurant well liked?"). Here is a brief discussion on it: x.com/ralphbrooks/status/17448400338723… Here is more on KTO: * Paper: github.com/ContextualAI/HALOs/blob/main…... * Code: github.com/ContextualAI/HALOs