Stanford's UniEvo-VL Self-Distillation Lifts Qwen-image GenEval From 0.747 to 0.808
stanfordnlp · hf · 2026-10-01
Stanford NLP introduces UniEvo-VL, an on-policy self-distillation recipe where one multimodal model acts as both teacher (seeing its own critique as privileged info) and student, minimizing divergence between their diffusion distributions. Built on open-source Qwen-image-2512, it lifts GenEval from 0.747 to 0.808 and GenEval2 Soft-TIFA from 32.97 to 35.53. Stronger external critics (e.g., GPT5.6-Luna) suggest a higher self-evolution ceiling, and gains are uneven across tasks like text rendering.
More from Multimodal
- Seedream 5.5 night portraits go viral for film-like lighting and reflections — SimplyAnnisa · 2026-11-15
- 'Into the Frame': a Marble-powered browser game jumping across 17,000 years of art history — msharmas · 2026-11-15
- MiniMax H3 image editing tested locally: 6 steps, ~50s per edit on RTX 4070S — Ant_6431 · 2026-11-15
- New ChatGPT image failure mode: faces warp on handstand poses — flowersslop · 2026-11-15
- Fable 5.2 composes a piano piece 'Elsewhere, Then Here' and animates its own video — Antique_Cat4404 · 2026-11-15
- Creator claims no AI video platform makes money: Sora shut down, Grok Imagine layoffs — AIandDesign · 2026-11-15