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 16 of 59

May 11, 2026SonoWorld and the Semantic Prerequisites of Spatial AudioCurrent 3D scene generation methods can transform a single photograph into an explorable virtual environment. You can walk through gardens, cross bridges, and inspect waterfalls from any angle. Yet these worlds remain ee May 11, 2026When Cameras Hallucinate: Rethinking Image Authenticity in the Age of Generative ISPsThe image signal processor (ISP) has long operated as an opaque intermediary between silicon and screen. Every photograph taken on a modern smartphone undergoes extensive computational transformation within this speciali May 11, 2026The Understanding-Generation Gap: Why LLMs Verify Better Than They Generate, and How to Fix It Introduction There is a peculiar asymmetry at the heart of modern multimodal AI systems. When presented with an image and asked whether it contains four apples or five, a large language model will reliably count correct May 10, 2026The Illusion of Multimodal Understanding: Why Large Models Fail at Genuine Visual Reasoning in STEMRecent advances in Multimodal Large Language Models (MLLMs) have created widespread optimism regarding their capacity to reason across text and vision. Yet beneath the surface of benchmark saturation lies a troubling rea May 10, 2026 May 10, 2026Hierarchical Planning with Latent World Models: Reconsidering Temporal Abstraction for Zero-Shot Robotics Introduction Model Predictive Control (MPC) with learned world models represents one of the most promising pathways toward generalist robotic systems. By learning predictive dynamics from experience and simulating futur May 10, 2026The Preference Gap: Why Mobile Agents Fail at True PersonalizationMobile GUI agents have reached an inflection point. Benchmarks like AndroidWorld and MobileWorld demonstrate that contemporary models can navigate complex multi-app workflows, execute precise touch gestures, and complete May 10, 2026Learning in Context Space: When Prompts Become Parameters The Optimization Geometry of Agent Memory The machine learning community has spent decades refining how to optimize neural network weights, yet we are only beginning to understand how to optimize what might be called th May 10, 2026 May 10, 2026Sample Routing as a Hybrid Strategy for Reinforcement Learning in Language ModelsPost-training of large language models through Reinforcement Learning with Verifiable Rewards has emerged as the dominant paradigm for enhancing reasoning capabilities in mathematical, scientific, and agentic domains. Wh May 10, 2026Improving Code Generation Through Unsupervised Self-Distillation: The Precision-Exploration ConflictThe relentless pursuit of stronger code generation capabilities has traditionally relied on an expanding apparatus of external supervision. Reinforcement learning demands carefully engineered reward models and verifiers; May 10, 2026Feature Economics and the Illusion of Robustness in Phishing Detection Introduction The machine learning May 10, 2026Verifier-Backed Hard Problem Generation: Breaking the Reward Hacking Cycle in Mathematical Self-PlayThe asymmetry between solving and generating mathematical problems has emerged as a critical bottleneck in the development of autoregressive language models. While contemporary systems demonstrate PhD level performance o May 10, 2026Decomposing Musical Structure in the Frequency Domain: An Analysis of the Latent Fourier Transform Introduction Modern generative music models operate through hierarchical decomposition. Whether through diffusion processes that denoise signals from coarse to fine representations, or autoregressive models that predict May 10, 2026From Atomic Mapping to World Models: Reimagining Visual Generation as Cognitive ArchitectureThe current generation of visual synthesis systems, from Midjourney to Stable Diffusion 3 and Flux, has achieved a level of photorealism that renders the Turing Test for images effectively obsolete. Yet this technical tr May 9, 2026Strategic Trajectory Abstraction: Decoupling Planning from Execution in Agentic Reinforcement LearningLarge language models have demonstrated impressive capabilities as interactive agents, yet their deployment on long-horizon decision making tasks remains fraught with challenges. The core difficulty lies not in generatin May 9, 2026Repetition over Diversity: Why Quality Trumps Quantity in Non-English Language Model TrainingThe dominant narrative in large language model development has long been governed by scaling laws: more parameters, more compute, and crucially, more data. This framework, articulated by Kaplan et al. and refined by Hoff May 9, 2026Bridging Learning and Adaptation: Hybrid RL-Extremum Seeking for Robust Robotic ManipulationThe deployment of deep reinforcement learning in physical robotic systems has long been hindered by a fundamental brittleness. Policies trained to master specific manipulation tasks often degrade catastrophically when co