Signals and Queries: Token-Efficient Agentic Context Discovery
Read the original on arXiv Computation and Language →The paper introduces Signals and Queries (S&Q), a lightweight first-stage relevance filter for LLM agents that removes at least 30% of candidate policies before downstream evaluation while maintaining high recall. S&Q uses offline-generated activating examples and hard negatives to train a logistic classifier that predicts policy relevance via embeddings and vector dot products, eliminating per-turn LLM calls. Evaluated on 9,144 relevance labels across 254 conversation prefixes and 36 airline policies, S&Q outperforms top‑K and cosine‑threshold retrieval in precision at every recall level, consistently across two embedding providers.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.