Researchers from UIUC and Stanford propose RecursiveMAS, a multi-agent framework that replaces text-to-text communication with latent embedding passing. Instead of generating reasoning tokens at every step, agents loop continuous representations through RecursiveLink modules and only output text at the end. Tests across nine benchmarks show up to 2.4x faster inference, 75% token reduction by round three, and an 8.3% accuracy gain, with far cheaper training than full fine-tuning.
Swipe through stories, personalise your feed, and save articles for later — all on the app.
