← back to paper

Review history

arxiv: 2608.03357 · 2 revisions

Can Text-to-Image Models Draw from the Right Frame of Reference?

1. 2026-08-15 CONDITIONAL MODERATE v1.4.0-alltime-deepseek-medium novelty 6.0
   328014 ms 15202 in 43775 out 2026-08-15T14:54:15.000300+00:00
2. 2026-08-05 CONDITIONAL MODERATE v1.3.0-daily-deepseek novelty 7.0
   123905 ms 15174 in 12859 out 2026-08-05T00:59:28.965502+00:00