MODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning

In three linesMODF-SIR is a multi-agent framework built on a lightweight MLLM for social intelligence reasoning. It uses knowledge distillation, Test-Time Adaptation (TTA), and LoRA fine-tuning. Achieving state-of-the-art results with 30% of IntentTrain data across multiple benchmarks. Code, demo, and LoRA weights released.

Summary generated by Claude — human-verified