SAP open-sources STS: schema-free enterprise data synthesis hitting 100% constraint satisfaction
SAP · hf · 2026-10-10
SAP released and open-sourced Synthesis Through Simulation (STS), a schema-free paradigm for generating enterprise data to train and evaluate tool-calling agents.
- Core idea: an LLM agent generates data by executing operations against policy-enforcing APIs in simulated enterprise environments. Structural validity is guaranteed by construction, and validity enforcement is decoupled from distribution modeling
- The Generalist Populator (GP), a domain-agnostic agent, achieves 0.88 average marginal fidelity and 100% constraint satisfaction across ten environments without ever seeing DB schemas
- Baselines fall short: statistical synthesizers can't run on seven environments due to seed-data requirements, and schema-privileged agents fail 82% of trajectories on tightly coupled airline workflows
The full framework, all ten environments, and generated datasets are open-sourced at github.com/SAP/synthesis-through-simulation.
More from Research
- MathEconLib open-sourced: Axiom Math and top economists bring machine-checked proofs to economics in Lean 4 — soumitrashukla9 · 2026-10-23
- Ken Ono invites the community to build MathEcon on the new Lean 4 economics library — soumitrashukla9 · 2026-10-23
- IFM open-sources RL360: 8,192 concurrent sandboxes and ~1.3M rollouts for agentic RL — rupspace · 2026-10-23
- Science lifts expression of concern on 2024 AI debunking-conspiracies paper; conclusions unchanged — DG_Rand · 2026-10-23
- Embodied First Exam benchmarks coding agents on 1,090 tasks across 44 robots — ManlingLi_ · 2026-10-23
- Anima Anandkumar on neural operators: AI weather models beat supercomputers — AnimaAnandkumar · 2026-10-23