Unverified50% confidenceFactExact time
τ-bench is an agent evaluation benchmark proposed by the Sierra team, used to measure a model's policy compliance in "tool-agent-user" three-party interaction scenarios, covering two domains: airline customer service and retail e-commerce customer service.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAgentBench is a next-generation benchmark for LLM Agents that evaluates model performance across multiple dimensions including tool-use accuracy, multi-turn dialogue coherence, and cross-scenario generalization ability.65% similarUnverifiedTerminal-Bench 属于交互式Agent基准测试家族，与SWE-bench、WebArena等评测共同构成AI Agent能力的多维评估体系64% similarUnverifiedAgentic Benchmarks are designed to evaluate the autonomous performance of models in open, long-horizon, multi-objective environments.63% similarUnverifiedOn Product Hunt, Sierra categorizes Multimodal Agents under two categories: Customer Success and Artificial Intelligence63% similarUnverifiedEach task in τ-bench comes with a domain policy document specifying the conditions for refunds, rebooking, and exchanges; the agent must satisfy the user's requests across multiple turns without violating policy; evaluation metrics include task success rate and policy violation rate.62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1007498API
curl https://kongchang.com/api/v1/knowledge/claims/1007498MCP
get_claim(id=1007498)