Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note2 min read

Rehearsed Multi-Agent Live Product Demonstrations with Real-Time Voice Question Answering

Source

arXiv

arxiv.org/abs/2606.30294v1

Read the source

Source

Plain-Language Brief

The authors propose Rhetor, a multi-agent system that automates live product demonstrations by merging UI exploration and source-code analysis, producing rehearsed demos with synchronized narration and real-time voice question answering. The study reports the internal locator-firing rate of Rhetor on four applications and suggests a benchmark protocol for further evaluation.

What the Document Says

  • The authors present Rhetor, a multi-agent system that takes a running web application and its source-code repository as input and produces rehearsed live demonstrations with synchronized narration and real-time voice question answering.
  • Rhetor consists of cross-modal feature representation, grounded scripter, pre-presentation rehearsal loop, runtime synchronization invariant, and locator-firing rate.
  • The study reports the internal locator-firing rate (sigma-bar) on six pipeline sessions for four deployed applications.
  • A benchmark protocol of ten metrics across six application categories is defined for further evaluation.

Key Concepts Mentioned

  • multi-agent system
  • live product demonstrations
  • web application
  • source-code repository
  • narration
  • real-time voice question answering
  • cross-modal feature representation
  • grounded scripter
  • pre-presentation rehearsal loop
  • runtime synchronization invariant
  • locator-firing rate
  • benchmark protocol

Claims Made by the Document

  • Rhetor automates live product demonstrations with synchronized narration and real-time voice question answering.
  • The internal locator-firing rate of Rhetor spans 0.31-1.00 over 147 scripted actions on four applications.

Evidence Used

  • The authors describe the architecture and functioning of Rhetor in detail.
  • Data on the internal locator-firing rate is provided for six pipeline sessions on four deployed applications.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Live product demonstrations are a recurring, high-cost activity in software organizations: a human presenter must select features, dispatch the corresponding interactions on a running application, narrate them coherently, and answer questions in real time. Existing automation addresses only fragments -- generalist browser agents target instruction-conditioned task completion, and demo-video tools produce fixed MP4 artifacts that cannot be questioned and silently break under interface drift. We propose Rhetor, a multi-agent system that takes a running web application and its source-code repository as input and produces a rehearsed live demonstration with segment-synchronized narration and real-time voice question answering. The architectural contributions are a cross-modal feature representation that merges UI exploration with source-code analysis into features tagged with discrete focus tiers, a grounded scripter constrained to UI elements observed during exploration and dispatched through multi-strategy semantic locators, a pre-presentation rehearsal loop with explicit convergence and graceful degradation to narration-only segments, and a runtime synchronization invariant that ties each browser action to the audio-end event of its narration segment. Across six pipeline sessions on four deployed applications -- including the public-domain whiteboard application Excalidraw -- the rehearser's internal locator-firing rate (sigma-bar) spans 0.31-1.00 over 147 scripted actions; on the substantial workload (53 actions, full tier differentiation), sigma-bar is approximately 0.92, and on the public-domain reference point the locator-repair step drives convergence to sigma-bar = 1.00 at iteration 2. We additionally define a benchmark protocol of ten metrics across six application categories that would establish, beyond the case study, whether each design choice contributes positively.