It's 2026 and still no adversarially robust model exists on CIFAR-10
scaling01 · x · 2026-11-23
A post points out that as of 2026, there is still no adversarially robust model on CIFAR-10 — decades into adversarial example research. A reply pushes back asking whether adversarial examples can always be found, touching the core open debate over whether robustness is achievable in principle. A concise snapshot of a long-unsolved deep learning security problem.
More from Research
- RSIBench-Infra from ByteDance Seed and Tsinghua tests if agents can optimize production-scale AI infrastructure — geoffwolfe · 2026-11-23
- Autoresearch compresses the Riemann Hypothesis proof by 47% already — DynamicWebPaige · 2026-11-23
- Polymarket odds put Riemann Hypothesis at 77.8% as next Millennium Prize AI solves — Polymarket · 2026-11-23
- w2v-bert-fast open-sources exact 3x speedup for w2v-BERT 2.0 training and inference — kastnerkyle · 2026-11-23
- AI reads 2,000 historical illustrations: individual errors, near-perfect aggregate patterns — _akpiper · 2026-11-23
- Princeton/Meta evolve inference strategies: 39% cheaper Qwen3 runs, same accuracy — michaelTM_ai · 2026-11-23