3 ms·
When I want to solve a new problem with an agent, I always ask it to search broadly for prior work in the given area online, and then analyze if we can build ou
by jbergqvist 6mo ago
When I want to solve a new problem with an agent, I always ask it to search broadly for prior work in the given area online, and then analyze if we can build our solution using it as inspiration.
I see it as the solution being out there in “idea space”, and by having the agent search beforehand we can more efficiently explore this space before converging on the final solution.
- dalmo3 6mo agoIs it not safe to assume that all* publicly available prior work is in the training data? Then you could just prompt it to propose options with pros and cons etc. * Bar extremely new stuff from after the cutoff
- technotony 6mo agoMaybe but having it search first to load the context with relevant information sure gets better results
- SadErn 6mo ago[dead]
- formerly_proven 6mo agoIncluded in the training corpus doesn’t mean perfect or even partial recall.