Orchestra Research publishes skills for AI agents covering robot policy work with Physical Intelligence OpenPI models, including fine-tuning, checkpoint conversion and inference across ALOHA, DROID and LIBERO environments. The author's catalog also spans research writing for machine learning and systems venues, from paper drafting and citation checks to conference talk slides, along with model merging with mergekit and experiment tracking with SwanLab.
Ai Research Skills
19 skills · 14,319 stars · MIT
Academic Plotting Generates publication-quality figures for ML papers from research context.
Ara Compiler Compiles any research input — PDF papers, GitHub repositories, experiment logs, code directories,…
Ara Research Manager Records research provenance as a post-task epilogue, scanning conversation history at the end of…
Ara Rigor Reviewer Performs ARA Seal Level 2 semantic epistemic review on Agent-Native Research Artifacts, scoring s…
Autoresearch Orchestrates end-to-end autonomous AI research projects using a two-loop architecture.
Brainstorming Research Ideas Guides researchers through structured ideation frameworks to discover high-impact research direct…
Build conference talk slides from a paper in PowerPoint and DeckCraft Slides for oral, spotlight and invited talks, with speaker notes, as PDF and editable PPTX
Creative Thinking For Research Applies cognitive science frameworks for creative thinking to CS and AI research ideation.
Evaluating Cosmos Policy Evaluates NVIDIA Cosmos Policy on LIBERO and RoboCasa simulation environments.
Evolving Ai Agents Provides guidance for automatically evolving and optimizing AI agents across any domain using LLM…
Experiment Tracking Swanlab Provides guidance for experiment tracking with SwanLab.
Fine Tuning Openvla Oft Fine-tunes and evaluates OpenVLA-OFT and OpenVLA-OFT+ policies for robot action generation with c…
Fine Tuning Serving Openpi Fine-tune and serve Physical Intelligence OpenPI models (pi0, pi0-fast, pi0.5) using JAX or PyTor…
Ml Paper Writing Write publication-ready ML/AI papers for NeurIPS, ICML, ICLR, ACL, AAAI, COLM.
Ml Training Recipes Battle-tested PyTorch training recipes for all domains — LLMs, vision, diffusion, medical imaging…
Model Merging Merge multiple fine-tuned models using mergekit to combine capabilities without retraining.
Prompt Guard Meta's 86M prompt injection and jailbreak detector.
Pytorch Fsdp2 Adds PyTorch FSDP2 (fully_shard) to training scripts with correct init, sharding, mixed precision…
Systems Paper Writing Comprehensive guide for writing systems papers targeting OSDI, SOSP, ASPLOS, NSDI, and EuroSys.