Training an open-source 7B multimodal LLM with GRPO and a continuous reward, using only coarse quality scores, produces human-aligned interpretable text-to-image evaluation scores and rationales.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
T2I-Eval-R1: Reinforcement Learning-Driven Reasoning for Interpretable Text-to-Image Evaluation
Training an open-source 7B multimodal LLM with GRPO and a continuous reward, using only coarse quality scores, produces human-aligned interpretable text-to-image evaluation scores and rationales.