[1]
T. Gu, “UniME-V2: MLLM-as-a-Judge for Universal Multimodal Embedding Learning”, AAAI, vol. 40, no. 26, pp. 21378–21386, Mar. 2026.