11–20 of 29 posts
Any ideas how to solve the agent's don't have total common sense problem? I have found when using agents to verify agents, that the agent might observe something that a human would immediately find off-putting and obviously wrong but does not raise any flags for the smart-but-dumb agent.
Architecturally focusing on Episodic memory with feedback system.
This training is retrieved next time when something similar happens
Any ideas how to solve the agent's don't have total common sense problem? I have found when using agents to verify agents, that the agent might observe something that a human would immediately find off-putting and obviously wrong but does not raise any flags for the smart-but-dumb agent.
Only solution is to train the issue for the next time. Architecturally focusing on Episodic memory with feedback system. This training is retrieved next time when something similar happens
Testing voice agents would require some kind of knowledge integration. Do you have any plans to support custom knowledge bases for test voice agents ?
Moreover, we even generate scenarios from the knowledge base
How do you handle sessions where the correct outcome is an incomplete flow — e.g. the agent correctly refuses to move forwards because the caller failed verification, or correctly escalates to a human?
If we miss some cases, there's always a feedback loop to help improve your test suite
Earlier quoted context omitted.
Only solution is to train the issue for the next time. Architecturally focusing on Episodic memory with feedback system. This training is retrieved next time when something similar happens
Training is an overkill at this point imo. I have seen agents work quite well with a feedback loop, some tools and prompt optimisation. Are you doing fine-tuning on the models when you say training?
https://github.com/rush86999/atom/blob/main/docs/EPISODIC_ME...