François Chollet Defends Test-Time Training as 'Pure' Deep Learning vs Neurosymbolic Tradition
teortaxesTex · x · 2026-08-14
Keras founder François Chollet recently argued that test-time training is the only form of test-time adaptation that is "pure" deep learning. According to Chollet, it adapts in a continuous latent space rather than adapting in a discrete symbol space like neurosymbolic approaches.
This sparked a debate over foundational AI paradigms. A researcher pushed back, claiming that LLMs autoregressively generating discrete symbols was a victory for the neurosymbolic tradition all along. They argued this vindicates early symbolic AI pioneers like Newell & Simon, joking that "Hinton is destroyed" and "Minsky is proven right again."
More from Fun
- Vibe Coder Records Cross-Language Podcast via AI Translation — StewartalsopIII · 2026-08-14
- Anthropic's Multi-Agent Experiment Triggers a 'Turf War' on Shared Tasks — RebeccaBellan · 2026-08-14
- DeepSeek Hires Expert, Category Theory Appears in Harness Design — burny_tech · 2026-08-14
- AI Avatar Music Video: Psychedelic Cyberwave Anime — VIV-AF-D · 2026-08-14
- Engineer criticizes LLMs for overcomplicating tasks, like patching pants with titanium alloy — bingxu_ · 2026-08-14
- Developer Jokes About Using Shock Collars to Control Runaway API Inference Costs — curious_vii · 2026-08-14