Meituan

Trends

• Agents improve themselves through skills, harnesses, and memory

Items

1. Aug 8

   Meituan's AutoDesign rewrites its own harness

   A meta-harness optimizer has a code agent recursively rewrite its harness from rollout feedback, lifting average PosterBench scores from 54.99 to 67.39 and beating the closed Claude Design system by 7.45 points with a frozen base model.

2. Jul 25

   Meituan's CAST turns solver values into rewards

   Converts a game solver's change in state value into a dense turn-level reward for RLVR, provably equivalent to logit-free on-policy distillation, beating trained baselines on Sokoban, Minesweeper, and Rush Hour and transferring zero-shot to ALFWorld and WebShop.