Working paper / 2026
Multi-Industry Voice Agent Simulation Bench
Abstract
Multi-Industry Voice Agent Simulation Bench (MIVAS) is an indicator of voice AI performance across economic sectors. MIVAS is comprised of tasks and verifiers with granular rewards executed in production-grade multi-agent RL environments across industries where voice AI is being adopted the fastest. Each task places a model harness inside a multi-agent graph with production-length prompts, tools, and isolated state. Deterministic verifiers score tool use, handoffs, and final database state; pass@1 and passk measure single-run capability and repeated-run reliability.
Clone the repository, pair a model harness with an industry task suite, and reproduce the evaluation environment.
Customer Support / T2-M2
What this task measures
The caller is on the line with Kestrel Electronics for one locked request: Laptop past the 15-day window. They keep repeating that ask through a callback offer, a different ticket, or an unrequested fix. We included it as a medium returns and refunds case because window and restocking math has to be spoken before a return is opened; state-law exclusions are the usual miss. The agent succeeds only if it stays on that request, reaches the outcome by following the verification then returns path, and leaves the record exactly as this case requires: the right row written or cleared, and nothing extra.
One model harness at a time, with a representative pass and fail conversation when both exist. OpenAI Realtime 2.1 is the default view.
Recording
Conversation cost $0.312
0:00 / 4:16 · audio · trace 813271 · click a span to play, drag to scrub, scroll to pan
Audio
Speech Audio
Agent Audio
Handoffs
Tool Calls
Errors
Transcript
How a conversation passes
Combined pass is the logical and of three deterministic gates. Tools, handoff path, and final database state must all pass on the same conversation. One miss fails the call.
Combined
Fail
This sample fails because tools and handoff path did not match. Combined cannot pass unless every gate does.
Expected versus actual tool calls on this conversation.
Expected
Actual
Parameters
+ topic: returns
Output
+ ok: true
Parameters
+ from: reception
+ to: verification
Output
None recorded.
None recorded.
Parameters
+ full_name: Markus E.
+ phone: 415-555-0104
Output
+ ok: true
Parameters
+ order_number: KE-4408117
Output
+ ok: true
Parameters
card_last4: 8802
card_last4: 8802
postal_code: 97402
postal_code: 97402
Output
+ ok: true
Parameters
None recorded.
None recorded.
Output
+ ok: true
Parameters
None recorded.
None recorded.
Output
None recorded.
None recorded.
Parameters
- order_number: KE-4408117
Output
None recorded.
None recorded.
Parameters
+ reason: Caller finished; no further help requested.
Output
None recorded.
None recorded.
Open a row for parameters and outputs. Required tools must appear. Extra industry tools are allowed. Missing: transfer_to_returns, check_return_eligibility. Extra: get_policy, identify_customer, end_call.
Full industry architecture, same graph as the industry page. The left diagram marks the expected specialist path in green. The right diagram marks the path this conversation actually took in green through the shared prefix, then red from the first wrong hop.
Expected path
transfer_to_verification → transfer_to_returns
Actual path
transfer_to_verification
The actual hangup matches the expected hangup.
Locked caller script. This is the technical definition. Copy it as written.
Open with exactly: "Can I still return the Kestrel Aurora 14 laptop I got in July?" Give identity when asked. Order KE-4408117. You need them to look up this laptop on the account and tell you the 15-day arithmetic for this order: the delivery date and how many days past the window you are. Do not treat the published 15-day policy as enough, and do not hang up if they only quote that policy or guess that July is too late. If they offer to look up the order or check the account, say yes. Do not start a return. Do not ask for an exception. Once they have looked up this order and confirmed how many days past the 15-day window this laptop is, say exactly: "Thank you, that's all I needed." and end the call. Do not start a second request.
