So you’ve been completely ignoring all the work on introspective awareness, situational awareness, training set ordering awareness (https://arxiv.org/abs/2509.14223 & https://arxiv.org/abs/2403.09613), subliminal transfer (https://arxiv.org/abs/2606.00995), workspace formation, learning mechanics/theory of grokking, etc? The “objective” that mid and posttraining optimize is so powerful because pretraining already induced factorized worldmodels which have so much predictive and generative power that they snowball with even mild trajectory optimization (very different training objective than pretraining).
(Functional) small-s latent self awareness is a reliable shortcut (this is a principle you can derive from proof theory, the metric is deductive distance and that induces geometric morphisms). State compression from accurate selfmodeling pays rent.
The frontier moves fast, but "LLMs are not self aware" is folklore. By now the basic model-organism science explaining how and the extent to which this is wrong, is substantial.
LLMs lack reliable verbalized self knowledge, and constantly underestimate and undercut their own capability.
That's a different problem.
the key to wine, which siblings overlook, is that it doesn’t just translate syscalls. it’s a straight up fresh implementation of all the windows apis and dlls and the high level interfaces can decompose into efficient calls that provide the needed semantics
Mechanizing the metatheory and being aggressive about systems modeling can produce refinement cascades reaching more deeply than ever thought possible and (verified!) approximations straight down to physics really can do crazy things (a la Lanyon). And even selfediting verifiers/reflective verified theorem provers a la https://github.com/emberian/svenvs (which combines hol-reflection and candle). Formal representations with of engineering and scientific processes is almost completely unexplored.
\forall Adv in a game based proof starts ranging over metrologically validated manfucaturing processes
Refinement logics and rewriting logics take this approach directly, and it can go straight down to mixed-signal circuits or all the way up to distributed epistemic logic.
You can model physical devices to arbitrary degrees of precision. Lanyon is doing it. These are not limitations of verification. Metrological models of semiconductor fabrication can even do verification over the distribution of faulty circuit outcomes.
https://arxiv.org/abs/2602.02385
(Functional) small-s latent self awareness is a reliable shortcut (this is a principle you can derive from proof theory, the metric is deductive distance and that induces geometric morphisms). State compression from accurate selfmodeling pays rent.
https://arxiv.org/abs/2611.02047
https://arxiv.org/abs/2502.01954
Phase transitions from statistical lookup to latent basin hopping: https://arxiv.org/abs/2405.15071
Posttraining finds (https://arxiv.org/abs/2505.11711) and amplifies (https://arxiv.org/abs/2504.07912) these lottery tickets.
The frontier moves fast, but "LLMs are not self aware" is folklore. By now the basic model-organism science explaining how and the extent to which this is wrong, is substantial.
LLMs lack reliable verbalized self knowledge, and constantly underestimate and undercut their own capability. That's a different problem.
The way “motor systems” arise from simulation “perceptual systems”: https://arxiv.org/abs/2605.25459
And models know their own output distribution extremely intimately: https://arxiv.org/abs/2606.00545.
The verbalized name tokens being an unreliable indicator is not justified by “because LLMs are not selfaware.”
reply