MADS: Multi-Agent Dialog Simulation for Robust Conversational Agents
We present MADS (Multi-Agent Dialog Simulation), a synthetic data generation pipeline that produces structured, diverse dialogue trajectories.
ProofPaper ↗
Key points
- Task-oriented conversational agents remain fragile under realistic conditions, especially when users exhibit non-cooperative behavior.
- Existing function-calling benchmarks underrepresent adversarial conversation trajectories, thereby limiting the training resources available for developing robust agents.
- We inject adversarial conversation, producing validated conversation trajectories that capture realistic, unhappy paths.
- Using MADS, we construct MADS-Bench: a benchmark spanning six domains, 55 personas, and 1095 conversations, with approximately 66% of conversations containing failure-prone conversation scenarios.
Sources (1)
- [1]MADS: Multi-Agent Dialog Simulation for Robust Conversational AgentsarXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 10:08 PM
We present MADS (Multi-Agent Dialog Simulation), a synthetic data generation pipeline that produces structured, diverse dialogue trajectories.
Task-oriented conversational agents remain fragile under realistic conditions, especially when users exhibit non-cooperative behavior.
Extractive summary: sentences quoted from the sources.
Before this
- Oct 8, 2026StereoEngine: Scaling Stereo Bimanual Robot Data without Stereo Cameras
- Oct 8, 2026Is Real-World Training Data Necessary for Generalist Graph Anomaly Detection?
- Oct 8, 2026Conditional Transfer from Controlled Pretraining Mixtures to Code
- Oct 8, 2026VAMR: Multi-Question Agentic Reasoning for Efficient Long-Form Video Understanding
- Oct 8, 2026When Interfaces Speak: Data-Aware Generative UI Harness for Active Interaction
- Oct 7, 2026RECAST: Learning to Compute the Right Context through Adaptive Evidence Routing