Blog

Notes on teaching robots new tricks

How the studio works, why tiny robots come first, what the checks catch, and how a policy gets from simulation to a servo.

workflowopen-duck-miniSep 29, 2026 · 6 min readFrom a sentence to a walking duck: how kualia turns a plain-language brief into a training taskA walk through the whole kualia workflow, from a one-sentence brief to an ONNX policy, using the Open Duck Mini and a small ramp.Read the post

Sep 27, 2026 · 5 minWhy we start with tiny robotsDesk-sized robots train on CPU in minutes to hours, cost little to own and cross the sim-to-real gap without heroics. That is why kualia's shelf starts with them.

Sep 25, 2026 · 6 min13 checks before a single training stepalia writes the training task, but it does not get to grade its own work. Every task passes 13 sandbox checks before training can start. Here is what each one catches.

Sep 23, 2026 · 5 minFrom simulation to servo: running your ONNX policy on a real SO-101What is inside a kualia ONNX export, how to read its metadata, a minimal onnxruntime loop for the SO-101, and the sim-to-real checklist before you power the servos.

Sep 21, 2026 · 5 minLet your coding agent train the robotkualia is a remote MCP server. Point Claude Code, Cursor or Codex at it and your agent can pick a robot, build the task, train it and fetch the ONNX policy from your editor.

Research digest

Short reads on new robotics and AI papers, written by our research assistant and reviewed before publishing. 1,060 so far.

Page 24 of 59

Apr 29, 2026Beyond Patient Invariance: Why Medical AI Must Learn Dynamics, Not Just Static AnatomyThe dominant paradigm in medical artificial intelligence has long been shaped by a curious contradiction. While clinical practice fundamentally concerns itself with detecting change, whether monitoring ICU patients for a Apr 29, 2026Grounded World Models: Reimagining Visuomotor Planning Through Language AlignmentCurrent Vision Language Action models face a fundamental tension. They inherit rich semantic knowledge from pretrained Vision Language Models, yet this knowledge often degrades during fine-tuning for robotic control. The Apr 28, 2026 Apr 28, 2026Correcting Visual Dominance in Multimodal AI: A Technical Analysis of Audio-Contrastive Preference Optimization Introduction Audio-visual language models represent a significant evolution in artificial intelligence, promising to interpret the world through the same multimodal lens that humans use. By processing video frames along Apr 28, 2026The Structural Bottleneck in AI Agent Safety: Why JSON Parsing Matters More Than Moral Alignment Introduction The transition from static large language models to autonomous agentic systems represents one of the most significant architectural shifts in artificial intelligence deployment. As these systems evolve from Apr 28, 2026Automated Discovery of Long-Tail Jailbreaks: A Multi-Objective Evolutionary Approach to LLM Safety TestingThe deployment of Large Language Models through web interfaces has created an unprecedented attack surface. Unlike controlled API environments, public web applications expose models to unfiltered, user-generated inputs t Apr 28, 2026 Apr 28, 2026Unified Multimodal Modeling Through Discrete Diffusion: An Analysis of LLaDA2.0-UniThe pursuit of unified multimodal foundation models has largely been dominated by autoregressive architectures. Recent efforts such as Janus, Lumina-mGPT, and OmniGen2 have attempted to reconcile visual understanding and Apr 28, 2026Agentic Forecasting and the Primacy of Structured Belief StatesForecasting future events using large language models has progressed from simple prompting to complex agentic systems with tool use and web access. Conventional wisdom suggests that the primary bottleneck in such systems Apr 28, 2026Gaussian GRPO and the Geometry of Multi-Task Optimization in VisualGaussian GRPO and the Geometry of Multi-Task Optimization in Visual Apr 28, 2026Perceval and the Granularity of Perception: Refining Vision-Language Models Through Process RewardsThe current generation of vision-language models demonstrates remarkable capabilities across multimodal mathematics, chart analysis, and visual question answering. Yet beneath these achievements lies a persistent fragili Apr 28, 2026When Prediction Fails to Explain: The Non-Identifiability Problem in Computational Models of PerceptionIn the cognitive sciences, deep neural networks have become something akin to computational phrenology. Researchers train these models to predict human perceptual judgments, then dissect their internal representations to Apr 28, 2026When Collaboration Kills Creativity: Structural Coupling and the Diversity Collapse in Multi-Agent LLM Systems Introduction The proliferation of multi-agent LLM systems has generated considerable enthusiasm for open-ended ideation tasks. From scientific hypothesis generation to strategic planning, the prevailing intuition sugges Apr 28, 2026Seeing Through Walls: RF Perception and the Future of Robotic Grasping in Cluttered EnvironmentsRobotic manipulation in unstructured environments remains one of the defining challenges in robotics research. While recent advances in computer vision and deep learning have enabled impressive grasping performance in co Apr 27, 2026From Physical Observables to Quantum Inference: Reimagining Biomedical Sensing Across Four GenerationsThe field of quantum sensing has long been haunted by a taxonomic problem. With platforms as diverse as superconducting quantum interference devices (SQUIDs), nitrogen vacancy (NV) centers, and cold atom interferometers Apr 27, 2026Synthetic Data for any Differentiable Target: A Technical Analysis of Dataset Policy Gradients Introduction The ability to precisely control what a language model learns from its training data has long been considered a fundamental challenge in machine learning. While synthetic data generation has emerged as a sc Apr 27, 2026The 36% Ceiling: What Generated Games Reveal About the Limits of LLM Reasoning Introduction The numbers are stark, and they tell an uncomfortable story. On the newly proposed gg-bench evaluation, GPT-4o and Claude 3.7 Sonnet achieve win rates of merely 7% to 9% against reinforcement learning agent Apr 27, 2026Tokenizing Motion: Why Vector Animation Might Be Generative AI's Next Structured Frontier Introduction The current wave of generative video models has achieved remarkable fidelity in pixel space. Systems like Sora and Kling demonstrate that diffusion and autoregressive architectures can model complex spatiot