TabAgent:一种用表格-文本分类器替代代理生成组件的框架
TabAgent: Distilling LLM Agent Components into Cheap Classifiers
AI总结:
TabAgent通过表格-文本分类器替代生成决策组件,显著降低代理系统延迟和成本,提升任务成功率。
摘要:
代理系统,即能够自主执行多步骤工作流以实现复杂目标的AI架构,通常通过重复调用大型语言模型(LLM)来完成封闭集决策任务,如路由、短名单筛选、门控和验证。尽管方便,这种设计导致部署缓慢且昂贵,因为累积延迟和令牌使用量。我们提出了TabAgent,一种用于在封闭集选择任务中用在执行轨迹上训练的紧凑型文本-表格分类器替代生成决策组件的框架。TabAgent(i)从轨迹中提取结构化模式、状态和依赖特征(TabSchema),(ii)通过与模式对齐的合成监督增强覆盖范围(TabSynth),(iii)用轻量级分类器评分候选者(TabHead)。在长周期AppWorld基准测试中,TabAgent在保持任务级成功率的同时消除了短名单时间的LLM调用,将延迟减少了约95%,推理成本减少了85-91%。除了工具短名单筛选外,TabAgent还能推广到其他代理决策头,建立了一种学习判别替换生成瓶颈的范式,用于生产代理架构。
英文摘要:
Agentic systems, AI architectures that autonomously execute multi-step workflows to achieve complex goals, are often built using repeated large language model (LLM) calls for closed-set decision tasks such as routing, shortlisting, gating, and verification. While convenient, this design makes deployments slow and expensive due to cumulative latency and token usage. We propose TabAgent, a framework for replacing generative decision components in closed-set selection tasks with a compact textual-tabular classifier trained on execution traces. TabAgent (i) extracts structured schema, state, and dependency features from trajectories (TabSchema), (ii) augments coverage with schema-aligned synthetic supervision (TabSynth), and (iii) scores candidates with a lightweight classifier (TabHead). On the long-horizon AppWorld benchmark, TabAgent maintains task-level success while eliminating shortlist-time LLM calls, reducing latency by approximately 95% and inference cost by 85-91%. Beyond tool shortlisting, TabAgent generalizes to other agentic decision heads, establishing a paradigm for learned discriminative replacements of generative bottlenecks in production agent architectures.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。