ETH Zurich's PMAE Tackles MAE Masking Dilemma Beyond Pixel Space
kastnerkyle · x · 2026-08-02
ETH Zurich introduces PMAE to solve the double-edged sword of random patch masking in MAEs. Traditional pixel-space masking faces a dilemma: masking too much foreground destroys the learning signal, while masking redundant background allows the model to cheat by simply copying pixels. PMAE addresses this by moving beyond pixel-space masking.
Related event: ETH Zurich Introduces PMAE to Improve Masked Autoencoders(2 posts)→
More from Research
- Netflix details its production LLM judge: hundreds of thousands of recommendations scored weekly — omarsar0 · 2026-08-24
- Nature Comment: Provenance, not interpretability, grounds trust in autonomous science — gabepgomes · 2026-08-24
- New Architecture RHEA: Train 1B Model on 8GB VRAM — zemondza · 2026-08-24
- Trained two 16M-param models to do generative CAD with real physics — debreuil · 2026-08-24
- Claude model helps discover complex structure on S^6, solving 60-year-old math problem — Singularitarian · 2026-08-24
- Study: Agents read instructions/notes 60.5% of the time, rarely touch API docs — dair_ai · 2026-08-24