| CARVIEW |
Mathias Lechner
I am Co-founder and CTO of Liquid AI, a startup founded in 2023 to focus on building efficient foundation models at every scale. We raised a Series A led by AMD, Shopify, Samsung, and G42 at a valuation of $2.3B. We build and release state-of-the-art language models called LFM2 at 1B parameters and below that run efficiently on CPU and edge devices. LFM2 outperform similar sized models such as Llama3 and Qwen3 at both speed and performance.
I am also a Research Affiliate at the Computer Science and Artificial Intelligence Laboratory (CSAIL) at MIT working with Prof. Daniela Rus. I completed my PhD (2022) at the Institute of Science and Technology Austria (ISTA) supervised by Tom Henzinger. Prior to my PhD, I received my master (2017) and bachelor (2016) degrees in Computer Science at the Vienna University of Technology (TU Wien).
Together with my colleague Djordje Zikelic I received the 2023 Outstanding Scientific Achievement Award at the Institute of Science and Technology Austria for our work on proving safety in stochastic machine learning systems. With my colleague Ramin Hasani, I got the Hyperion Research 2022 HPC Innovation Excellence Award for my work on Liquid neural and was honored as Outstanding Reviewer at ICRA 2021. I co-led the team that won the F1-Tenth - Autonomous Racing Grand Prix at IFAC 2020, and I received the Distinguished Young Alumni award at TU Vienna in 2018 for my MS thesis.
Contact: mlechner (at) mit (dot) edu
Last update: July 11, 2025
Publications
-
Continuous Autoregressive Generation with Mixture of Gaussians
Alex Quach, Tsun-Hsuan Wang, Ramin Hasani, Mathias Lechner, Alexander Amini
ICML 2025 Workshop ES-FoMo-III, 2025
-
NeuralStoc: A Tool for Neural Stochastic Control and Verification
Matin Ansaripour, Krishnendu Chatterjee, Thomas A. Henzinger, Mathias Lechner, Abhinav Verma, Đorđe Žikelić
International Conference on Computer Aided Verification (CAV) - Tool Paper, 2025
-
SafeDiffuser: Safe Planning with Diffusion Probabilistic Models
Wei Xiao, Tsun-Hsuan Wang, Chuang Gan, Ramin Hasani, Mathias Lechner, Daniela Rus
International Conference on Learning Representations (ICLR) - Poster, 2025
-
Holistic Surgical Phase Recognition with Hierarchical Input Dependent State Space Models
H Wu, TH Wang, M Lechner, R Hasani, JA Eckhoff, P Pak, OR Meireles, ...
arXiv preprint arXiv:2506.21330, 2025
-
Competitive Multi-Team Behavior in Dynamic Flight Scenarios
T Seyde, M Lechner, J Rountree, D Rus
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024
-
Large scale dataset distillation with domain shift
N Loo, A Maalouf, R Hasani, M Lechner, A Amini, D Rus
International Conference on Machine Learning (ICML), 2024
-
Overparametrization helps offline-to-online generalization of closed-loop control from pixels
M Lechner, R Hasani, A Amini, TH Wang, TA Henzinger, D Rus
IEEE International Conference on Robotics and Automation (ICRA), 2774-2782, 2024
-
Learning with Chemical versus Electrical Synapses Does it Make a Difference?
M Farsang, M Lechner, D Lung, R Hasani, D Rus, R Grosu
IEEE International Conference on Robotics and Automation (ICRA), 15106-..., 2024
-
State-free inference of state-space models: The transfer function approach
RN Parnichkun, S Massaroli, A Moro, JTH Smith, R Hasani, M Lechner, ...
arXiv preprint arXiv:2405.06147, 2024
-
Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control
N Tumma, M Lechner, N Loo, R Hasani, D Rus
International Conference on Learning Representations (ICLR), 2024
-
Gigastep - One Billion Steps per Second Multi-agent Reinforcement LearningMathias Lechner, Lianhao Yin, Tim Seyde, Tsun-Hsuan Wang, Wei Xiao, Ramin Hasani, Joshua Rountree, Daniela Rus
Conference on Neural Information Processing Systems (NeurIPS) - Dataset and Benchmark Track, 2023
-
Compositional Policy Learning in Stochastic Control Systems with Formal GuaranteesĐorđe Žikelić*, Mathias Lechner*, Abhinav Verma, Krishnendu Chatterjee, Thomas A Henzinger
Conference on Neural Information Processing Systems (NeurIPS), 2023
-
On the Size and Approximation Error of Distilled DatasetsAlaa Maalouf, Murad Tukan, Noel Loo, Ramin Hasani, Mathias Lechner Daniela Rus
Conference on Neural Information Processing Systems (NeurIPS), 2023
-
Robust flight navigation out of distribution with liquid neural networksMakram Chahine*, Ramin Hasani*, Patrick Kao*, Aaron Ray, Ryan Shubert, Mathias Lechner, Alexander Amini, Daniela Rus
Science Robotics, 2023
-
Dataset Distillation with Convexified Implicit GradientsNoel Loo, Ramin Hasani, Mathias Lechner, Daniela Rus
International Conference on Machine Learning (ICML), 2023
-
On the Forward Invariance of Neural ODEsWei Xiao, Tsun-Hsuan Wang, Ramin Hasani, Mathias Lechner, Yutong Ban, Chuang Gan, Daniela Rus
International Conference on Machine Learning (ICML), 2023
-
A Learner-Verifier Framework for Neural Network Controllers and Certificates of Stochastic SystemsKrishnendu Chatterjee, Thomas A Henzinger, Mathias Lechner, Đorđe Žikelić
International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS), 2023
-
Liquid Structural State-Space ModelsRamin Hasani*, Mathias Lechner*, Tsun-Hsuan Wang, Makram Chahine, Alexander Amini, Daniela Rus
International Conference on Learning Representations (ICLR), 2023
-
Infrastructure-based End-to-End Learning and Prevention of Driver FailureNoam Buckman, Shiva Sreeram, Mathias Lechner, Yutong Ban, Ramin Hasani, Sertac Karaman, Daniela Rus
IEEE International Conference on Robotics and Automation (ICRA), 2023
-
Revisiting the Adversarial Robustness-Accuracy Tradeoff in Robot LearningMathias Lechner, Alexander Amini, Daniela Rus, Thomas A. Henzinger
IEEE Robotics and Automation Letters (RA-L), 2023
-
Quantization-aware Interval Bound Propagation for Training Certifiably Robust Quantized Neural NetworksMathias Lechner, Đorđe Žikelić, Krishnendu Chatterjee, Thomas A. Henzinger, Daniela Rus
AAAI Conference on Artificial Intelligence (AAAI), 2023
-
Learning Control Policies for Stochastic Systems with Reach-avoid GuaranteesĐorđe Žikelić*, Mathias Lechner*, Thomas A. Henzinger, Krishnendu Chatterjee
AAAI Conference on Artificial Intelligence (AAAI), 2023
-
Closed-form continuous-time neural networksRamin Hasani*, Mathias Lechner*, Alexander Amini, Lucas Liebenwein, Aaron Ray, Max Tschaikowski, Gerald Teschl, Daniela Rus
Nature Machine Intelligence, 2022
-
PyHopper-A Plug-and-Play Hyperparameter Optimization EngineMathias Lechner, Ramin Hasani, Sophie Neubauer, Philipp Neubauer, Daniela Rus
Has it Trained Yet? NeurIPS 2022 Workshop, 2022
-
Are All Vision Models Created Equal? A Study of the Open-Loop to Closed-Loop Causality GapMathias Lechner, Ramin Hasani, Alexander Amini, Tsun-Hsuan Wang, Thomas A Henzinger, Daniela Rus
NeurIPS 2022 Machine Learning for Autonomous Driving Workshop (ML4AD), 2022
-
Mixed-Memory RNNs for Learning Long-term Dependencies in Irregularly-sampled Time SeriesMathias Lechner, Ramin Hasani
Memory in Artificial and Real Intelligence (MemARI) NeurIPS 2022 Workshop, 2022
-
Infinite Time Horizon Safety of Bayesian Neural NetworksMathias Lechner*, Đorđe Žikelić*, Krishnendu Chatterjee, Thomas A. Henzinger
Conference on Neural Information Processing Systems (NeurIPS), 2021
-
Causal Navigation by Continuous-time Neural NetworksCharles J Vorbach*, Ramin Hasani*, Alexander Amini, Mathias Lechner, Daniela Rus
Conference on Neural Information Processing Systems (NeurIPS), 2021
-
On-Off Center-Surround Receptive Fields for Accurate and Robust ImageZahra Babaiee, Ramin Hasani, Mathias Lechner, Daniela Rus, Radu Grosu
International Conference on Machine Learning (ICML), 2021
-
Adversarial Training is Not Ready for Robot LearningMathias Lechner, Ramin Hasani, Radu Grosu, Daniela Rus, Thomas A. Henzinger
IEEE International Conference on Robotics and Automation (ICRA), 2021
-
Scalable Verification of Quantized Neural NetworksThomas A. Henzinger*, Mathias Lechner*, and Djordje Zikelic* (alphabetical)
AAAI Conference on Artificial Intelligence (AAAI), 2021
-
Liquid Time-constant NetworksRamin Hasani*, Mathias Lechner*, Alexander Amini, Daniela Rus, and Radu Grosu
AAAI Conference on Artificial Intelligence (AAAI), 2021
-
On the Verification of Neural ODEs with Stochastic GuaranteesSophie Grünbacher, Ramin Hasani, Mathias Lechner, Jacek Cyranka, Scott A. Smolka, and Radu Grosu
AAAI Conference on Artificial Intelligence (AAAI), 2021
-
Neural circuit policies enabling auditable autonomyMathias Lechner*, Ramin Hasani*, Alexander Amini, Thomas A. Henzinger, Daniela Rus, and Radu Grosu
Nature Machine Intelligence, 2020
-
The Natural Lottery Ticket Winner: Reinforcement Learning with Ordinary Neural CircuitsRamin Hasani*, Mathias Lechner*, Alexander Amini, Daniela Rus, and Radu Grosu
International Conference on Machine Learning (ICML), 2020
-
Learning Representations for binary classification without backpropagationMathias Lechner
International Conference on Learning Representations (ICLR), 2020
-
An SMT Theory of Fixed-Point ArithmeticMarek Baranowski, Shaobo He, Mathias Lechner, Thanh Son Nguyen, and Zvonimir Rakamaric
International Joint Conference on Automated Reasoning (IJCAR), 2020
-
Gershgorin Loss Stabilizes the Recurrent Neural Network Compartment of an End-To-End Robot Learning SchemeMathias Lechner*, Ramin Hasani*, Daniela Rus, and Radu Grosu
IEEE International Conference on Robotics and Automation (ICRA), 2020
-
How Many Bits Does it Take to Quantize Your Neural Network?Mirco Giacobbe*, Thomas A. Henzinger*, and Mathias Lechner* (alphabetical)
International Conference on Tools and Algorithms for the Construction and Analysis of Systems (TACAS), 2020
-
Lagrangian Reachtubes: The Next GenerationSophie Gruenbacher, Jacek Cyranka, Mathias Lechner, Md. Ariful Islam, Scott Smolka, and Radu Grosu
IEEE Conference on Decision and Control (CDC), 2020
-
Response characterization for auditing cell dynamics in long short-term memory networksRamin Hasani*, Alexander Amini*, Mathias Lechner, Felix Naser, Radu Grosu, and Daniela Rus
International Joint Conference on Neural Networks (IJCNN), 2019
-
Designing worm-inspired neural networks for interpretable robotic controlMathias Lechner*, Ramin Hasani*, Manuel Zimmer, Thomas A. Henzinger, and Radu Grosu
IEEE International Conference on Robotics and Automation (ICRA), 2019