FortisAVQA and MAVEN: a Benchmark Dataset and Debiasing Framework for Robust Multimodal Reasoning Paper โข 2504.00487 โข Published 20 days ago โข 18
ฯ-Decoding: Adaptive Foresight Sampling for Balanced Inference-Time Exploration and Exploitation Paper โข 2503.13288 โข Published Mar 17 โข 50