202503_trig

One paper was accepted by Reasoning and Planning for LLMs @ ICLR2025!
Towards visual text grounding of multimodal large language model.