Researchers question viral TDV paper: compared against nerfed baselines, doesn't scale
y_m_asano · x · 2026-11-07
In a discussion around the viral TDV paper ("You Don't Need Strong Assumptions"), @gabriberton argues its comparisons used a nerfed iBOT and DINO rather than the real DINO — and the paper even loses to those weakened baselines. He adds that TDV doesn't scale to larger datasets, while crediting the authors for being honest about this. @ymasano suggests continuing the thread with Platonic representation papers.
Related event: Viral TDV Paper Questioned Over Weakened Baselines and Scalability(3 posts)→
More from Research
- Google: stale TODO comments derail 1 in 7 coding agent sessions — rseroter · 2026-11-07
- JHU paper: treating demographic annotation disagreement as the target for valid LLM annotation — krisgligoric · 2026-11-07
- VLM video annotation shortcuts: highest-accuracy configs can yield wrong conclusions, JHU study finds — krisgligoric · 2026-11-07
- Applied Compute shares recipe for production RL with continual learning — ypatil125 · 2026-11-07
- Muir linear attention streams 10M tokens in 25MB state, trains 2.7x faster than Transformer at 32K — _akhaliq · 2026-11-07
- ICLR slammed for poor score discrimination: 4-point scale squeezes papers into one narrow band — TuhinChakr · 2026-11-07