OSL and Neuralese: reading edition

Download original file

This is the original TeX manuscript source. Use the paper’s PDF for typeset equations.

1 \bibitem[CC77]{cousot} Patrick Cousot and Radhia Cousot. Abstract interpretation: a unified lattice model for static analysis of programs by construction or approximation of fixpoints. \emph{POPL}, 238--252, 1977. Author-maintained bibliographic record: \url{https://www.di.ens.fr/~cousot/COUSOTpapers/POPL77.shtml}.
2 \bibitem[N97]{necula} George C. Necula. Proof-carrying code. \emph{POPL}, 106--119, 1997. DOI: \href{https://doi.org/10.1145/263699.263712}{10.1145/263699.263712}.
3 \bibitem[HYC08]{session} Kohei Honda, Nobuko Yoshida, and Marco Carbone. Multiparty asynchronous session types. \emph{POPL}, 273--284, 2008. DOI: \href{https://doi.org/10.1145/1328438.1328472}{10.1145/1328438.1328472}.
4 \bibitem[G+23]{causal} Atticus Geiger et al. Causal Abstraction: A Theoretical Foundation for Mechanistic Interpretability. \emph{arXiv:2301.04709}, initially 2023, versioned research manuscript. \url{https://arxiv.org/abs/2301.04709}.
5 \bibitem[H+24]{coconut} Shibo Hao, Sainbayar Sukhbaatar, DiJia Su, Xian Li, Zhiting Hu, Jason Weston, and Yuandong Tian. Training Large Language Models to Reason in a Continuous Latent Space. \emph{arXiv:2412.06769}, initially 2024. \url{https://arxiv.org/abs/2412.06769}. Cited for the latent-recurrence proposal, not as a result obtained in this release.
6 \bibitem[S+25]{dilemma} Denis Sutter, Julian Minder, Thomas Hofmann, and Tiago Pimentel. The Non-Linear Representation Dilemma: Is Causal Abstraction Enough for Mechanistic Interpretability? \emph{arXiv:2507.08802}, 2025. \url{https://arxiv.org/abs/2507.08802}.

Back to source controls

Back to OSL and Neuralese: reading edition