Benedikt Stroebl

I'm the CTO and cofounder of Ludus Labs. Before that, I was a PhD student at Princeton University, advised by Arvind Narayanan.

At Ludus Labs, we're building physically-simulated environments for AI agents to compete in sports-like competitions, creating a new form of entertainment while advancing AI capabilities.

My research focuses on AI agents, with a focus on enhancing their real-world usefulness and reliability. Part of that is developing rigorous evaluation frameworks and studying the limitations of inference scaling techniques.

Selected Publications & Projects

• HAL: The Holistic Agent Leaderboard for Centralized and Reproducible Agent Evaluation Benedikt Stroebl*, Sayash Kapoor*, Arvind Narayanan (2025)

• Localized Cultural Knowledge is Conserved and Controllable in Large Language Models Veniamin Veselovsky*, Berke Argin*, Benedikt Stroebl*, Chris Wendler, Robert West, James Evans, Thomas L. Griffiths, Arvind Narayanan (2025)

• Inference Scaling 𝙛Laws: The Limits of LLM Resampling with Imperfect Verifiers Benedikt Stroebl*, Sayash Kapoor, Arvind Narayanan arXiv preprint 2411.17501 (2024)

• CORE-Bench: Fostering the Credibility of Published Research Through a Computational Reproducibility Agent Benchmark Zachary S. Siegel, Sayash Kapoor, Nitya Nagdir, Benedikt Stroebl, Arvind Narayanan arXiv preprint 2409.11363 (2024)

• AI Agents That Matter Sayash Kapoor*, Benedikt Stroebl*, Zachary S. Siegel, Nitya Nadgir, Arvind Narayanan arXiv preprint 2407.01502 (2024)

(* indicates equal contribution)

Blog Posts

• Is AI progress slowing down? Making sense of recent technology trends and claims. Arvind Narayanan, Benedikt Stroebl, Sayash Kapoor AI Snake Oil (2024)

• AI leaderboards are no longer useful. It's time to switch to Pareto curves. Sayash Kapoor*, Benedikt Stroebl*, Arvind Narayanan AI Snake Oil (2024)

Workshops

• Workshop on Useful and Reliable AI Agents Princeton University. 600+ attendees. Virtual Workshop. August 2024.

Talks & Press

• Building and evaluating AI agents that matter. AWS Applied Scientists. Invited talk. March 2025.

• AI agents that matter. Weaviate Podcast. Podcast. September 2024.

• AI agent benchmarks are misleading, study warns. VentureBeat. News article. June 2024.

• The perils of evaluating AI agents. Meta (Core Applied Sciences). Invited talk. May 2024.