MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?
Paper • 2407.04842 • Published • 55
Multimodal
Note Base for discussion about NSFW and Biases in AI
MMEvalPro: Calibrating Multimodal Benchmarks Towards Trustworthy and Efficient Evaluation
Paper • 2407.00468 • Published • 36Note Critique on why some benchmarks are biased when comparing LLM and MLLM. New benchmark fixing those errors
Qwen2 Technical Report
Paper • 2407.10671 • Published • 175Note New SOTA - better than llama!