Post

What happens when we can't use recursive belief to compete? We can use anomaly detection instead! Here, we (led by soon-to-be-Dr @nitalon.bsky.social ) devise a multi-agent account where compression & reward expectation are used to notice deception jair.org/index.php/ja...

In recursive modeling (I think that you think...), agents eventually hit a cognitive limit. It’s a logical trap - humans can't just "think harder" forever. It might also be the case that humans infrequently want to avoid using recursive strategies when they can -

With this, more equitable social outcomes are achieved, and exploitation is reduced when compression & expectation are tuned within normal bounds. It provides a plausible, mechanistic formal framework in which agents can compete without recursion.

The model offers another multi-agent route to paranoia alongside hypermentalising. If the anomaly detector is tuned too tightly, it yields false +ves. The agent becomes overly defensive against benign peers. This connects generalist accounts of delusions to social contexts.