Lists (4)
Starred repositories
Online Self-Distillation from Real-Time Feedback for Computer-Use Agents https://arxiv.org/abs/2609.40253
AutoGUIWorld: Image Generators as Visual World Models for GUI Agent
WebAnything: From the Wild Web to Scalable Agent Environments
Scalable pipeline for synthesizing verifiable RLVR training data for computer-use agents
Agent World Model in the Wild: Computer-Use Agent Environment Synthesis
Agent self-evolution evaluation — not static snapshots, but longitudinal growth curves. Measures transfer efficiency, error avoidance, and skill-hit quality through controlled experiments with and …
PrismaCoder: Pushing the Boundaries of Open Multimodal Code Intelligence
🌟 A curated list of papers, methods, and resources on long-horizon credit assignment for agentic RL.
[ICML 2026] Milestone-Guided Policy Learning for Long-Horizon Language Agents
Dataset of hackable TerminalBench-style tasks and exploit trajectories
8000
The official code of Agent-World: Scaling Real-World Environment Synthesis for Evolving General Agent Intelligence
The official implementation of "EnvScaler: Scaling Tool-Interactive Environments for LLM Agent via Programmatic Synthesis".
[AAAI 2026] Official codebase for "GenPRM: Scaling Test-Time Compute of Process Reward Models via Generative Reasoning".
Code for Trajectory-Level Speculation with Semantic Verification for Tool-Using Agents
Exploring Multi-Session Agentic Reinforcement Learning
AutoEnvScaling: Automating the Data Flywheel with Terminal Agents
Designing effective multimodal environment distributions beyond simply environment scaling.
RSIH — versionable, shareable agent harness: Pi coding agent + Genome config layer
Official code for "From Table to Cell: Attention for Better Reasoning with TabAlign" (NeurIPS 2026)
The offical repo for "Dream-RSI: Recursive Self-Improvement through Evolving Worlds"
[NeurIPS 2026] Beyond Data Scaling: Representation-Centric Continued Pre-training for Vision-Language-Action Models
[Neurips 2026] Official repository for the paper "Composition-RL: Compose Your Verifiable Prompts for Reinforcement Learning of Large Language Models"
Qwen-CUA: Native Computer Use for (Almost) Everything — a screenshot-driven agent that operates computers with keyboard and mouse, jointly developed by the Qwen Team and XLang Lab.