Ander Artola Velasco

Max Planck Institute for Software Systems, Germany.

prof_pic.jpg

Welcome! I’m Ander, a PhD. student at the Max Planck Institute for Software Systems, under the supervision of Dr. Manuel Gomez Rodriguez and as part of the IMPRS-TRUST program. Previously, I obtained my masters in Physics at Heidelberg University under the supervision of Dr. Sarah Bosman. I graduated in 2023 from the Complutense University of Madrid in Mathematics and Physics.

Some of my (past) interests include:

  • Machine learning, uncertainty quantification, explainability and support systems
  • Bayesian modelling and inference
  • Causality
  • Decision-making and strategic behavior
  • (Differential geometry and its applications)
  • (Cosmology, dark matter, the intergalactic medium and reionisation)
  • Anything that is intensive in abstract reasoning: from coding to philosophy

You can find my CV here.

news

Nov 15, 2025 I will give a spotlight talk at the PAIG workshop at EurIPS’25!
Oct 19, 2025 I gave a research talk at AICS (AI, Computing and Society) on our papers on LLM pricing and auditing!
Oct 17, 2025 We have recently released a new preprint! In the context of sequential decision-making, we develop a decision support system that uses a pre-trained AI agent to narrow down the set of actions a human can take to a subset, and then asks the human to take an action from this action set. Along the way, we also develop a best-arm identification algorithm for Lipschitz bandits suited to our problem. This work has done with an amazing team at MPI-SWS: Eleni Straitouri, Stratis Tsirtsis and Manuel Gomez Rodriguez.
Oct 08, 2025 We’ve released a new preprint! In today’s LLM-as-a-service ecosystem, users are vulnerable to providers overcharging them by misreporting the number of tokens required to encode the generated string. We develop a sequential auditing framework that enables a trusted third-party auditor to detect whether a provider is misreporting tokens. Our framework offers strong theoretical guarantees, and in experiments, it detects a misreporting provider after at most 70 queries.
Jun 26, 2025 I will be attending ICML’25 to present our paper Is Your LLM Overcharging You? Tokenization, Transparency, and Incentives at the Tokenization Workshop!