MIBE introduces a multi-subject interaction benchmark (MIB) with silver and gold sets and a dual-head evaluator (MIE) trained on VLM labels that outperforms baselines in matching human judgments.
Multibanana: A challenging benchmark for multi-reference text-to-image generation
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Introduces OmniRef-Bench benchmark and DyRef two-stage framework using Difficulty-aware Advantage Reweighting and Discriminative Reward Scaling to improve open-source models on complex multi-reference image generation.
citing papers explorer
-
MIBE: Multi-subject Interaction Benchmark and Evaluator for Personalized Image Generation
MIBE introduces a multi-subject interaction benchmark (MIB) with silver and gold sets and a dual-head evaluator (MIE) trained on VLM labels that outperforms baselines in matching human judgments.
-
Scaling Multi-Reference Image Generation with Dynamic Reward Optimization
Introduces OmniRef-Bench benchmark and DyRef two-stage framework using Difficulty-aware Advantage Reweighting and Discriminative Reward Scaling to improve open-source models on complex multi-reference image generation.