$τ$-Knowledge: Evaluating Conversational Agents over Unstructured Knowledge
This work addresses a gap in realistic, fully agentic evaluation over unstructured data in long-horizon interactions for developers of conversational AI agents.
This paper introduces $τ$-Knowledge, a benchmark for evaluating conversational agents that must retrieve and apply domain-specific knowledge from unstructured corpora during live interactions. In the $τ$-Banking domain, agents achieve only ~25.5% pass rate, indicating significant challenges in retrieving correct documents and reasoning over complex policies.
Conversational agents are increasingly deployed in knowledge-intensive settings, where correct behavior depends on retrieving and applying domain-specific knowledge from large, proprietary, and unstructured corpora during live interactions with users. Yet most existing benchmarks evaluate retrieval or tool use independently of each other, creating a gap in realistic, fully agentic evaluation over unstructured data in long-horizon interactions. We introduce $τ$-Knowledge, an extension of $τ$-Bench for evaluating agents in environments where success depends on coordinating external, natural-language knowledge with tool outputs to produce verifiable, policy-compliant state changes. Our new domain, $τ$-Banking, models realistic fintech customer support workflows in which agents must navigate roughly 700 interconnected knowledge documents while executing tool-mediated account updates. Across embedding-based retrieval and terminal-based search, even frontier models with high reasoning budgets achieve only $\sim$25.5% pass^1, with reliability degrading sharply over repeated trials. Agents struggle to retrieve the correct documents from densely interlinked knowledge bases and to reason accurately over complex internal policies. Overall, $τ$-Knowledge provides a realistic testbed for developing agents that integrate unstructured knowledge in human-facing deployments.