T-DIG trains an LLM prompt policy with TPIPS reward, VQA and HPSv3 penalties
simi_97k · x · 2026-10-28
T-DIG's training recipe: an LLM policy generates diverse prompts for a base prompt, trained on image-derived signals — maximizing TPIPS diversity along the requested axis while penalizing deviations in VQA alignment score and HPSv3 human preference score.
Related event: CMU Open-Sources T-DIG: Text-Conditioned Diversity for Image Generation(7 posts)→
More from Multimodal
- AI band Techhalla reimagines Slipknot's 'Wait and Bleed' as a 1971 soul performance — Mother-Pear-2392 · 2026-10-28
- Inside the Hailuo H3 Max retro-MV workflow: 15-second chunks, one reference image, multi-shot prompts — techhalla · 2026-10-28
- Open-source pipeline turns Seedance AI video into real PBR textures on your 3D mesh — OdinLovis · 2026-10-28
- AI-generated cyberpunk short 'VHS Sirens' shows off consistent characters and neon city detail — brythehuman · 2026-10-28
- First-time creator makes full animated music video with 40 clips via MiniMax H3 — Capable-Insurance-58 · 2026-10-28
- Midjourney sref combo 138606468+1306896884 yields textured flat illustration — sergeantsref · 2026-10-28