@misc{indiciaea0bd09d91b3a,
  title = {BASIC: Boosting Visual Alignment with Intrinsic Refined Embeddings in Multimodal Large Language Models},
  author = {Jianting Tang and Yubo Wang and Haoyu Cao and Linli Xu},
  year = {2025},
  url = {https://arxiv.org/abs/2508.06895},
  note = {Source identifier: 2508.06895}
}