Dev critiques compiling fast programs into slow neural nets: fun, but not useful
ctjlewis · x · 2026-10-16
Developer ctjlewis weighed in on Cheng's experiment of "compiling" already-fast programs into neural networks, while taking aim at Will's hype.
His take: it's a fun idea, probably done before, but not very useful — nobody wants to turn fast programs into slow neural nets; it might work as a student exercise. A representative counterpoint to the neural-compilation approach.
More from Research
- Researchers derive a dropout scaling law from active parameter count — gordic_aleksa · 2026-10-16
- Gemini is a robotics outlier: dropping it lifts capability-robotics correlation from 0.20 to 0.96 — ycombinator · 2026-10-16
- EPFL team uses RFdiffusion to design de novo metamorphic proteins switching between two folds — ferruz_noelia · 2026-10-16
- Manipulating a model's sense of power may reveal its takeover counterfactuals — jd_pressman · 2026-10-16
- Poser benchmark unmasks alignment-faking LLMs, one detection method catches 98% — jd_pressman · 2026-10-16
- Six Years of Evidence That MLP Neurons Are an Interpretable Basis, in One Thread — aryaman2020 · 2026-10-16