[2507.12549] The Serial Scaling Hypothesis
Abstract:While machine learning has advanced through massive parallelization, we identify a critical blind spot: some problems are fundamentally sequential. These "inherently serial" problems-from mathematical reasoning to physical simulations to sequential decision-making-require sequentially dependent computational steps that cannot be efficiently parallelized. We formalize this distinction in complexity theory, and demonstrate that current parallel-centric architectures face fundamental limitations on such tasks. Then, we show for first time that diffusion models despite their sequential nature are incapable of solving inherently serial problems. We argue that recognizing the serial nature of computation holds profound implications on machine learning, model design, and hardware development.
Published as a conference paper at ICLR 2026 T HE S ERIAL S CALING H YPOTHESIS Yuxi Liu1∗ Konpat Preechakul1∗ Kananart Kuwaranancharoen2 Yutong Bai1 1 2 UC Berkeley Independent Researcher A BSTRACT While machine…
saved by
related reading
- [2507.12549] The Serial Scaling Hypothesisarxiv.org
- Building a Reasoning Machine - Ph.D. thesis | Edward Huedwardjhu.com
- Scaling is subtler than it seemsberen.io
- How To Scale Your Modeljax-ml.github.io
- As Rocks May Think | Eric Jangevjang.com
- On neural scaling and the quanta hypothesisericjmichaud.com
- The Extreme Inefficiency of RL for Frontier Models - Toby Ordtobyord.com
- (Some of) The Models, They Just Don't Want to Learn | Tildeblog.tilderesearch.com
- The Scaling Hypothesis · Gwern.netgwern.net
- Scaling Laws, Carefully | Lil'Loglilianweng.github.io
- machine learning imindslice.substack.com
- My picture of the present in AI — LessWronglesswrong.com