1
AnE: Pushing the Reasoning Frontier of Multimodal LLMs via Anchor Evolution
多模态大模型推理新突破，AnE方法通过锚点演化机制提升复杂推理能力，极具理论价值与实践潜力。
arXiv:2605.25571v1 Announce Type: new Abstract: Post-training via Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) is crucial for enhancin…