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 57 of 59

Mar 18, 2026Thinking to Recall: When Reasoning Becomes Memory RetrievalThe prevailing intuition about Chain of Thought reasoning in large language models suggests that its value lies in decomposition. When a model tackles a mathematical proof or a multi-hop question requiring logical connec Mar 16, 2026 · TRSürü Karar Dinamiklerinde Duygusal Modülasyon: Biyolojik İşaretlerden Dijital İzlemlere Giriş: Arı Sürülerinden İnsan Topluluklarına Kolektif karar verme mekanizmaları, merkezi bir kontrol otoritesi olmaksızın bireysel etkileşimlerden makro düzeyde fikir birliklerinin ortaya çıktığı evrensel bir süreçtir. Mar 16, 2026Emotional Modulation in Swarm Decision Dynamics: Bridging Biological Signaling and Digital Collective Intelligence Introduction Collective decision making in biological systems has long provided elegant mathematical metaphors for understanding human coordination. The honeybee swarm's nest-site selection process, formalized through t Mar 15, 2026 · TR Mar 15, 2026Beyond Prompt Engineering: Why Personalization in Explainable AI Requires Closed-Loop ControlThe integration of Large Language Models into Explainable Artificial Intelligence has created a seductive trap. Faced with the complexity of technical explanation artifacts, practitioners increasingly rely on LLMs to tra Mar 15, 2026 · TRBilimsel Disiplinlerde Görsel Düzenleme: GRADE Benchmark'ının Işığında Yapay Zeka Modellerinin Çıkarım Sınırları Giriş Birleşik çok modlu modellerin (Unified Multimodal Models, UMM'ler) son yıllardaki yükselişi, görsel algı ve üretim yeteneklerinin tek bir çatı altında toplanması yönünde önemli adımlar sundu. Bu modeller, doğal di Mar 15, 2026The Symbolic Gap in Unified Multimodal Models: What GRADE Reveals About Discipline-Informed Image EditingThe rapid convergence of multimodal understanding and generation into unified architectures has created powerful systems capable of following natural language instructions to edit images. Yet these models, despite their Mar 15, 2026 · TRCompute-Optimal Örnekleme: LLM RL Eğitiminde Hesaplama Bütçesinin Optimal Dağılımı Giriş Büyük dil modellerinin (LLM) ön eğitiminde (pre-training) geçerli olan ölçeklendirme yasaları (scaling laws) sayesinde, hesaplama bütçesi ile model performansı arasındaki ilişki öngörülebilir hale gelmiştir. Chinc Mar 15, 2026IsoCompute Playbook: Rethinking Resource Allocation for LLM Post TrainingThe era of large language model development has been defined by scaling laws. From the Chinchilla compute optimal training paradigm to the predictable emergence of capabilities with model size, these empirical regulariti Mar 15, 2026 · TRUzun ve Yüksek Çözünürlüklü Videolarda Verimli Görü Anlama: AutoGaze ve Ötesindeki Semantik Uçurum Giris Çok modlu büyük dil modelleri (Multi-modal Large Language Models, MLLMs), genel amaçlı video anlama alanında önemli ilerlemeler kaydetmiş olsa da, uzun süreli ve yüksek çözünürlüklü videolar karşısında temel bir v Mar 15, 2026Attend Before Attention: A Critical Analysis of Reconstruction Based Token Reduction in Video MLLMsThe computational demands of processing long, high resolution videos remain the primary bottleneck preventing widespread deployment of multi-modal large language models in real world applications. While recent architectu Mar 15, 2026 · TR Mar 15, 2026Beyond the Left-to-Right Bottleneck: Joint Distribution Modeling in Multimodal AI Introduction Multimodal large language models have typically relied on a familiar architectural pattern. A pretrained large language model serves as the cognitive backbone, augmented with perception modules for vision a Mar 15, 2026 · TRCerrahi Yapay Zekada Akıl Yürütme: SUREON ve Uzman Anlatımlarından Öğrenme Giriş Mevcut cerrahi yapay zeka sistemleri, görsel algılamada önemli ilerlemeler kaydetmiş olsa da, temel bir sınırlama ile karşı karşıya. Bu sistemler görmeyi başarıyla öğrenmişlerdir, ancak düşünmeyi henüz öğrenemedil Mar 15, 2026From Pattern Matching to Clinical Cognition: The SUREON Benchmark and Surgical ReasoningCurrent surgical artificial intelligence has reached a perceptual plateau. While modern systems reliably identify instruments, phases, and anatomical structures within the operative field, they remain largely incapable o Mar 14, 2026 · TRFiziksel Simülatör Döngüsünde Video Üretimi: Görsel Gerçekçilik ile Fiziksel Tutarlılığın Birleştirilmesi Giriş Son yıllarda diffusion model tabanlı video üretim sistemleri, görsel kalite açısından dikkate değer ilerlemeler kaydetti. Ancak bu modellerin ürettiği içerikler, çarpışma mekanikleri, kütle ve yerçekimi gibi temel Mar 14, 2026 Mar 14, 2026 · TR