AI Daybook
AI Daybook
오늘의 지면
논문에이전트

G-CAST, 변화하는 수요에 따라 전자상거래 에이전트를 지속 학습

G-CAST: Governed Continual Self-Training for E-Commerce Agents under Evolving Demands

매체
Hugging Face
게시
중요도
3/5
화제 지수
10

한국어 요약

전자상거래 환경에서 사용자 요구와 에이전트 도구·기능이 계속 변하는 상황에서, 일일 운영 데이터를 보상 라우팅·엔트로피 우선순위 학습 데이터로 변환해 에이전트를 지속 학습시키는 G-CAST 시스템을 제시했다. 보상 진화를 조절하고 누적 평가로 체크포인트 전환을 관리해 새로운 행동을 학습하면서도 기존 기능을 유지한다. 알리바바 1688 플랫폼의 B2B 조달 에이전트 Yuancai에 배포됐다.

원문 발췌 · English

With the rapid advancement of Large Language Models (LLMs), an increasing number of users are turning to LLM-powered agents for conversational shopping in e-commerce. However, the user needs continue to change as new product categories, procurement scenarios, and feature requests emerge. At the same time, the product's execution harness-including its tools, skills, and operating procedures-continues to evolve to support increasingly sophisticated interactions. These changes create two practical challenges: 1) Demand drift alters the distribution, intent, and complexity of user requests, causing a model trained once to fall behind; 2) Harness evolution requires the model to use newly introduced capabilities correctly without losing behavior that users already rely on. Periodic manual retraining is too delayed and labor-intensive for this setting. To address these challenges, we present G-CAST, a governed continual self-training system that converts daily production evidence under evolving demand and capability contracts into reward-routed, entropy-prioritized training data and governed reward evolution. Governed checkpoint transition under cumulative evaluation controls each model update, enabling the agent to learn newly introduced behaviors without sacrificing previously accepted capabilities. Experiments on a real production e-commerce agent show that G-CAST achieves stronger adaptation to newly introduced demand while better preserving previously accepted behavior compared with a strong autonomous post-training baseline. Ablations further show that both reward-routed, entropy-prioritized data orchestration and governed reward evolution are important to these gains. G-CAST is now deployed in Yuancai(http://cai.1688.com/), a B2B procurement agent on Alibaba's 1688 platform.

#agent#e-commerce#continual learning
원문 기사 읽기
전체 보기 →