News for hackers. Stuff that matters.
Re:Opus 4.7 to 4.6 Inflation is ~45% by AlexCoventry on Sunday April 19, 2026 @04:04AM (#47822595)
> In a subsequent round, it generally can't meaningfully introspect on its prior internal state

It has the K/V cache, no?

Re:Opus 4.7 to 4.6 Inflation is ~45% by Sinidir on Sunday April 19, 2026 @11:23AM (#47824978)
The K/V Cache is just an optimization. But yeah you would expect the attention for the model producing "Ok im doing X" and you asking "Why did you do X?" be similar. So i don't see a reason why introspection would be impossible. In fact trying to adapt a test skill where the agent would write a new test instead of adapting a new one i asked it why and it gave the reasoning it used. We then adapted the skill to specifically reject that reasoning and then it worked and the agent adapted the existing test instead.