← back to paper

Review history

arxiv: 2608.04448 · 2 revisions

When does training on downscaled images yield the same gradients?

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   133129 ms 31809 in 16029 out 2026-08-15T14:40:04.364509+00:00
2. 2026-08-06 CONDITIONAL MODERATE v1.4.0-daily-deepseek-medium novelty 7.0
   133413 ms 31809 in 13361 out 2026-08-06T23:54:05.846699+00:00