AGORA: Adapter-Grounded Observation-Action Retention for Inference-Free Prompt Compression in LLM Agents 文章

ArXiv CS.AI2026-05-27NEWSen作者: Haoran Zhang, Zhaohua Sun
查看原文 →
关系图谱

AGORA: Adapter-Grounded Observation-Action Retention for Inference-Free Prompt Compression in LLM Agents · 相关技术

相关技术

ORMadapter通用表达式语言远程代码执行（RCE）divide-and-conquer partitioningUCTTokenSSESPAPromptPPRLayout-as-PolicyLMMLLM agentsForDAPTANNAgent