Research | Liquid AI

1.12.2025

LFM2 Technical Report

Alexander Amini, Anna Banaszak, Harold Benoit, Arthur Böök, Tarek Dakhran, Song Duong, Alfred Eng, Fernando Fernandes, Marc Härkönen, Anne Harrington, Ramin Hasani, Saniya Karwa, Yuri Khrustalev, Maxime Labonne, Mathias Lechner, Valentine Lechner, Simon Lee, Zetian Li, Noel Loo, Jacob Marks, Edoardo Mosca, Samuel J. Paech, Paul Pak, Rom N. Parnichkun, Alex Quach, Ryan Rogers, Daniela Rus, Nayan Saxena, Bettina Schlager, Tim Seyde, Jimmy T.H. Smith, Aditya Tadimeti, Neehal Tumma

Machine Learning Done Right.

Explore Our Research.

LFM2 Technical Report

LFM2 Technical Report

LFM-1B-Math: Can Small Models Be Concise Reasoners?

LFM-1B-Math: Can Small Models Be Concise Reasoners?

Exploring Diffusion Transformer Designs via Grafting

Exploring Diffusion Transformer Designs via Grafting

How Effectively Does a Model Use Its Memory

How Effectively Does a Model Use Its Memory

Convolutional Multi-Hybrids for Edge Devices

Convolutional Multi-Hybrids for Edge Devices

Automated Architecture Synthesis via Targeted Evolution

Automated Architecture Synthesis via Targeted Evolution

From Liquid Neural Networks to Liquid Foundation Models

From Liquid Neural Networks to Liquid Foundation Models

Towards a theory of learning dynamics in deep state space models

Towards a theory of learning dynamics in deep state space models

The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof

The Empirical Impact of Neural Parameter Symmetries, or Lack Thereof

State-Free Inference of State-Space Models: The *Transfer Function* Approach

State-Free Inference of State-Space Models: The *Transfer Function* Approach

Exploring Diffusion Transformer Designs via Grafting

Exploring Diffusion Transformer Designs via Grafting

How Effectively Does a Model Use Its Memory

How Effectively Does a Model Use Its Memory

Convolutional Multi-Hybrids for Edge Devices

Convolutional Multi-Hybrids for Edge Devices

Automated Architecture Synthesis via Targeted Evolution

Automated Architecture Synthesis via Targeted Evolution

LFM-1B-Math: Can Small Models Be Concise Reasoners?

LFM-1B-Math: Can Small Models Be Concise Reasoners?

From Liquid Neural Networks to Liquid Foundation Models

From Liquid Neural Networks to Liquid Foundation Models

Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control

Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control

Leave-one-out Distinguishability in Machine Learning

Leave-one-out Distinguishability in Machine Learning

COCO-Periph: Bridging the Gap Between Human and Machine Perception in the Periphery

COCO-Periph: Bridging the Gap Between Human and Machine Perception in the Periphery

Understanding Reconstruction Attacks with the Neural Tangent Kernel and Dataset Distillation

Understanding Reconstruction Attacks with the Neural Tangent Kernel and Dataset Distillation

Graph Metanetworks for Processing Diverse Neural Architectures

Graph Metanetworks for Processing Diverse Neural Architectures

How Structured Data Guides Feature Learning: A Case Study of the Parity Problem

How Structured Data Guides Feature Learning: A Case Study of the Parity Problem

Drive Anywhere: Generalizable End-to-end Autonomous Driving with Multi-modal Foundation Models

Drive Anywhere: Generalizable End-to-end Autonomous Driving with Multi-modal Foundation Models

Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control

Leveraging Low-Rank and Sparse Recurrent Connectivity for Robust Closed-Loop Control

Exploring Modern Evolution Strategies in Portfolio Optimization

Exploring Modern Evolution Strategies in Portfolio Optimization

Gigastep - One Billion Steps per Second Multi-agent Reinforcement Learning

Gigastep - One Billion Steps per Second Multi-agent Reinforcement Learning

Power your business, workflows, and engineers with Liquid AI.

Manage your preferences

State-Free Inference of State-Space Models: The Transfer Function Approach

State-Free Inference of State-Space Models: The Transfer Function Approach