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 15 of 59

May 12, 2026Constrained Semantic Guidance: Reformulating Linguistic Negation as Geometric Feasibility in Diffusion ModelsCurrent vision-language models exhibit a peculiar blind spot. They can generate "a highway at sunset" with photorealistic fidelity, yet stumble when asked for "a highway at sunset with no vehicles." This failure reveals May 12, 2026The Binding Problem in Textual Space: Why LLMs Fail at Viewpoint Rotation Despite Encoding Correct Spatial Representations Introduction Spatial intelligence has long been considered a cornerstone of artificial general intelligence. While recent advances in vision language models have demonstrated impressive capabilities in visual spatial re May 12, 2026Cognitive Primitives in Modular Language Models: What Brainstacks Reveals About Fine-TuningWhen a medical question about drug interactions routes through chat and mathematics adapters to reach an optimal answer, we must reconsider what fine-tuning actually accomplishes. In Brainstacks: Cross-Domain Cognitive C May 12, 2026The Indexical Gap: Why Visual Grounding Must Embrace Hand Pointing in Egocentric Vision Introduction: The Ambiguity of "That One" When a person wearing AR glasses points toward a cluttered table and asks, "What is that?", the query contains minimal descriptive content. The word "that" carries no visual fea May 12, 2026The Modality Gap in Vision-Language Models: When Adding Images Hurts PerformanceOne of the central assumptions driving multimodal artificial intelligence is that integrating additional sensory channels, particularly vision, should enhance reasoning capabilities. Yet a rigorous new study challenges t May 12, 2026When Multi Physics Unification Meets Epistemic Uncertainty Introduction The fragmentation of deep learning approaches to partial differential equations has long been a bottleneck in scientific machine learning. Current architectures, whether Fourier Neural Operators, DeepONets, May 12, 2026When Reasoning Fails: Query Misalignment and the Limits of Self-Knowledge in Large Language Models Introduction The deployment of reasoning capable large language models has introduced a troubling paradox. Models that generate explicit step-by-step rationales through Chain-of-Thought (CoT) prompting demonstrate super May 12, 2026Simulating Reason: How Physics Engines Are Reshaping LLM Training DataThe emergence of reinforcement learning with verifiable rewards (RLVR) has catalyzed remarkable advances in large language model reasoning, exemplified by systems like DeepSeek-R1. Yet beneath these achievements lies a c May 12, 2026The Mean Versus The Individual: Why Reward Models Struggle with Personalization and What It Means for RLHFReinforcement Learning from Human Feedback (RLHF) has become the dominant paradigm for aligning Large Language Models with human values, yet a fundamental tension lies at its core. The methodology implicitly assumes that May 11, 2026The Recipe Matters More Than the Kitchen: Why AI Weather Models Systematically Fail at ExtremesOver the past four years, machine learning has fundamentally altered the landscape of numerical weather prediction. From GraphCast to Aurora, Pangu-Weather to FourCastNet, data-driven models now rival or exceed the skill May 11, 2026The Tension Between Safety and Adaptation in Continual Reinforcement LearningWhen we deploy reinforcement learning agents into the physical world, we ask them to solve two contradictory tasks simultaneously. They must remain rigid enough to preserve safety constraints that prevent damage to thems May 11, 2026Task-Adaptive Reasoning in Egocentric Vision: Why One Size Fails AllEgocentric video understanding represents one of the most demanding frontiers in computer vision, precisely because it violates the static assumptions that underlie most visual recognition systems. When a camera moves th May 11, 2026Governance as Architecture: How HAAS Reframes Human-AI Collaboration Through Policy-Aware Allocation Introduction The allocation of work between humans and artificial intelligence systems has long suffered from a categorical error. Organizational design often treats the decision as binary; a task belongs either to a hu May 11, 2026The Statistical Geometry of Fairness: Why Auditing Predictors Across Subgroups Requires Cubically More DataIn predictive modeling, calibration represents a minimal statistical hygiene requirement. A predictor is calibrated when, conditional on predicting probability $v$, the event indeed materializes with frequency $v$. This May 11, 2026The Reliability Paradox: Why Smaller Responsibilities Make Better On-Device AIThe push toward on-device AI has been framed as liberation from cloud dependency. Privacy guarantees, zero latency, and offline functionality represent a compelling vision for mobile applications. Yet the technical liter May 11, 2026The Geometry of Silence: Why Multi-Agent Systems Should Stop Talking to Each OtherMulti-agent systems built upon large language models have rapidly evolved from simple prompt-chaining architectures into sophisticated reasoning engines capable of solving mathematical olympiad problems and generating co May 11, 2026Mastery Before Brevity: Gated Compression as a Post-Training Paradigm for Efficient ReasoningLarge language models capable of chain-of-thought reasoning have unlocked remarkable problem-solving abilities, yet this capability arrives with a hidden tax. The tendency toward verbose, meandering reasoning paths, what May 11, 2026