arXiv cs.CL· Yam Marcovitz·· 13 小时前AI 评分31
Signals and Queries（S&Q）：面向 Parlant 4.0 的 Token 高效智能体上下文发现
Signals and Queries: Token-Efficient Agentic Context Discovery
AI 导读
为 Parlant 4.0 设计的轻量级首阶段相关性过滤器 Signals and Queries（S&Q）可在下游评估前剔除至少 30% 的候选策略，同时保持高召回率。
正文 · 原文
Abstract:User-facing LLM agents must identify which behavioral policies apply at each turn, but evaluating every policy with an LLM is costly. We introduce Signals and Queries (S&Q), a lightweight first-stage relevance filter developed for Parlant 4.0 that removes at least 30% of candidate policies before downstream evaluation while preserving high recall. For each policy, an LLM generates activating examples and hard negatives offline. An independent logistic classifier trained on their embeddings then predicts relevance using embeddings and vector dot products, with no per-turn LLM calls. This shifts expensive relevance evaluation to a smaller candidate set, while allowing each policy's classifier to be updated independently. Evaluated against 9,144 LLM-judge relevance labels covering 254 $\tau^2$-bench conversation prefixes and 36 airline policies, S&Q achieves higher precision than top-K and cosine-threshold retrieval at every evaluated recall level, with consistent results across two embedding providers. These findings support inexpensive, per-policy classification as a practical first stage for reducing downstream evaluation workload in customer-facing agents.
| Subjects: | Computation and Language (cs.CL) |
| Cite as: | arXiv:2610.29326 [cs.CL] |
| (or arXiv:2610.29326v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2610.29326 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yam Marcovitz [view email]
[v1]
Wed, 21 Oct 2026 07:25:59 UTC (108 KB)
来源：arXiv cs.CL · arxiv.org