"One bad apple can spoil the bunch π", and that's doubly true for language agents!
Our new paper shows how monitoring and intervention can prevent agents from going rogue, boosting performance by up to 20%. We're also releasing a new multi-agent environment π΅οΈββοΈ
Using our method, we see consistent gains across varying difficulty levels of WhoDunitEnv. We apply our method to GovSim, a recent resource sharing environment and show an increase in both survival rates and efficiency of agents.
@giorgiopiatti
1
4
114
We believe monitoring and intervention to be a strong tool when dealing with complex systems. Read more here arxiv.org/abs/2502.05986
@megamor2 @OriYoran
1
6
106
Ohav retweeted
Do you have a "tell" when you are about to lie?
We find that LLMs have βtellsβ in their internal representations which allow estimating how knowledgeable a model is about an entity π£π¦π§π°π³π¦ it generates even a single token.
Paper: arxiv.org/abs/2406.12673v1 π§΅
@dhgottesman



