Rollback the World, Keep the Reflection: Rollback-Induced Reflection for Long-Horizon LLM Agents
Large language model (LLM) agents increasingly tackle long-horizon tasks through multi-step environment interaction, yet a single erroneous action can alter subsequent states and observations, causing errors to compound over time. Existing methods either correct the context without repairing altered environment states...
Yi Yu, Liuyi Yao, Yaliang Li et al.
· 1 citation