$\tau$-Elicitation: Benchmarking multi-turn entity extraction in voice agents

arXiv cs.AISeptember 15, 2026en

arXiv:2609.13602v1 Announce Type: new Abstract: Voice agents often need to collect names, addresses, identifiers, dates, and times exactly, yet end-to-end benchmarks obscure where capture fails. We introduce $\tau$-Elicitation, a 200-task voice benchmark spanning 10 entity types, controlled difficulty, caller realisms, and three environments. A matched text agent passes all tasks, but four voice configurations achieve robust exact success from 0.14 to 0.41. Agents increase verification for hard and unfamiliar entities and sometimes for incorrect captures, but not for their weakest caller voice; only 24 to 37 percent of verified errors are repaired. A scaffold that prescribes spelling, read-b

This is a short summary published by AI Global Wire. The full article is owned and hosted by arXiv cs.AI — open it there to read it in full.

• Forskning
• Agenter