Csaba Szepesvári pushes back on LeCun's generalization framework: the extra structure determines the answer
CsabaSzepesvari · x · 2026-10-12
RL researcher Csaba Szepesvári responds to LeCun's self-supervised generalization framework with a methodological critique: such questions can only be answered once extra structure is assumed, and that structure fundamentally determines the answer — so what is there to learn? What is genuinely unexplained or surprising here?
More from Research
- IFM open-sources RL360: 8,192 concurrent sandboxes and ~1.3M rollouts for agentic RL — rupspace · 2026-10-23
- Carbon-X DNA model slammed: worse than random on 8/10 clinical variant cases — anshulkundaje · 2026-10-23
- Science lifts expression of concern on 2024 AI debunking-conspiracies paper; conclusions unchanged — DG_Rand · 2026-10-23
- Embodied First Exam benchmarks coding agents on 1,090 tasks across 44 robots — ManlingLi_ · 2026-10-23
- Anima Anandkumar on neural operators: AI weather models beat supercomputers — AnimaAnandkumar · 2026-10-23
- Scaling GLM-5.3-Flash training to 1M token context, 64x over Megatron baseline — AndrewLBeam · 2026-10-23