Chain of thought becomes more compressed after the model is aware of the token budget
AI summary
I think this might actually be an instance of improved CoT controllability in a real-world scenario—the more the model notices it is "running out of budget," the more compressed its thinking becomes @tomekkorbak @MicahCarroll
Full text · Original
I think this is actually might be an instance of improved CoT controllability in the wild -- the more model notices that it is "runs out of budget" more compressed the thinking becomes
@tomekkorbak @MicahCarroll
Our original attack allows extracting reasoning of the recent frontier models, including Astra and Sol 6.1.
On reasoning effort MAX both Astra and Sol become very aware of their "token budgets", and eventually start saving tokens by omitting white spaces
More examples on https://stolen-thoughts.com/View the quoted post on X