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 8 of 59

May 29, 2026Where Identity Lives: Rethinking MLLM Conditioning for Subject-Driven GenerationSubject-driven image generation sits at an awkward intersection. On one side, we want the model to faithfully reproduce a specific subject, the user's particular dog, backpack, or stuffed animal, down to its fine visual May 29, 2026Manufacturing 4D Supervision: How Reshoot-Anything Sidesteps the Multi-View Data ProblemCamera control is one of the most persistent unsolved problems in video generation. We can synthesize photorealistic frames, follow text prompts, and condition on poses, yet asking a model to take an existing dynamic vid May 29, 2026Encoding Agents as Simplex Vertices: A Closer Look at Gamma-WorldMost interactive video world models still imagine the world as a stage for one actor. You feed a single action stream, one keyboard input, or one camera trajectory, and the model rolls out a plausible future. But the env May 29, 2026When Algorithms Agree Too Much: The Asymmetry of Strategic Monoculture Introduction There is a recurring anxiety in the literature on AI in markets: as more economic decisions get delegated to systems built on the same handful of foundation models, those systems will start behaving alike. May 29, 2026Two Timescales of Learning: Why LLMs Should Adapt in Context and in Weights Introduction There is a recurring tension in how we adapt large language models to new tasks. On one side sits parameter updating: supervised finetuning, preference optimization, and reinforcement learning with verifiab May 29, 2026Why Vision-Language Models Forget What They See: Analyzing Persistent Visual MemoryWhen a large vision-language model (LVLM) generates a long answer about an image, something subtle and counterproductive happens inside its attention mechanism. The longer the model talks, the less it actually looks. Thi May 29, 2026Why Decision Transformers Cannot Stitch, and How Hierarchy Fixes ItDecision Transformer (DT) reframed offline reinforcement learning as a sequence modeling problem, and in doing so it imported a great deal of machinery from the language modeling world: causal transformers, autoregressiv May 29, 2026Why Reward Hacking Is a Geometry Problem: A Look at General Preference Reinforcement LearningThe post-training stack for large language models has quietly split into two camps that rarely talk to each other. On one side sits online reinforcement learning with verifiable rewards, the engine behind the reasoning g May 28, 2026Smart Grid Security Through Intelligent Feature Selection: When Less Data Yields Better DetectionThe intersection of cybersecurity and power systems has become one of the most critical challenges in modern infrastructure protection. As our electrical grids evolve into sophisticated cyber-physical systems, the tradit May 28, 2026EvoStruct: Solving the Vocabulary Collapse Problem in Antibody Design Through Evolutionary-Structural IntegrationThe field of computational antibody design has made remarkable strides in recent years, yet a persistent and fundamental problem has continued to plague even the most sophisticated approaches: vocabulary collapse. A new May 22, 2026The Privacy Illusion of Synthetic Tabular Data: Insights from the MIDST ChallengeSynthetic data has emerged as the default prescription for data privacy compliance. Under frameworks such as GDPR, PIPEDA, and CCPA, organizations increasingly view generative models as mechanisms to anonymize sensitive May 22, 2026Bridging High-Level Intent and Low-Level Control: A Technical Analysis of Vega's Approach to Instructional Autonomous DrivingThe integration of natural language into autonomous driving systems has progressed rapidly from simple navigational commands toward complex scene understanding. Yet a persistent gap remains between linguistic comprehensi May 22, 2026Repurposing 3D Diffusion for Spatial Intelligence: A Technical Analysis of LaviGenThe disconnect between linguistic competence and physical reasoning remains one of the most persistent challenges in artificial intelligence. While large language models can eloquently describe the arrangement of furnitu May 22, 2026Bridging the Conceptual Rift: Structured Commitment Tracking in Complex Image GenerationText-to-image generation has undergone a subtle but significant shift in objectives. Early systems prioritized visual plausibility; contemporary benchmarks demand precise intent realization. When a user requests an image May 22, 2026When Video Models Answer Correctly but Understand Nothing: A New Benchmark for Physical Grounding The Specter of Ungrounded Competence Large vision language models can now describe complex physical events with apparent sophistication. Ask a model about a video showing liquid pouring into a container, and it will lik May 22, 2026Reproducibility and the Promise of Meta-Learning in Continual Learning: An Analysis of La-MAML Introduction The field of Continual Learning (CL) remains one of the most stubborn challenges in artificial intelligence. Unlike the static, independent and identically distributed assumptions that underpin most deep le May 22, 2026DexGrasp-Zero: Rethinking Cross-Embodiment Transfer Through Morphology-Aligned Graph Representations Introduction The proliferation of dexterous robotic hands, from the Shadow Dexterous Hand to the LEAP Hand and Inspire Robotics platforms, presents both an opportunity and a challenge for the robotics community. While h May 21, 2026The Virtue of Imperfection: When Flawed Rewards Improve Reinforcement LearningThe prevailing wisdom in reinforcement learning from human feedback (RLHF) treats reward model errors as a kind of poison. The assumption is simple and intuitive; any deviation between the proxy reward and the ground tru