Topic
An RL environment defines the tasks, observations, actions, transitions, and outcomes through which a policy gains experience. For frontier AI, environment engineering increasingly includes browsers, code repositories, simulators, synthetic users, and tool APIs. Environment diversity and fidelity often determine whether gains transfer beyond a benchmark.
Task distributions
Curriculum design
Simulation
Environment fidelity
Latest research
Z.ai
An official GLM-5.3 release note describing a post-training-only update on the same base model as GLM-5.2, with expanded executable environments for complex coding, security, and long-horizon professional workflows.
Fiora Starlight
A LessWrong proposal to let policies report exploitable RLVR environment bugs after a rollout, reward high-quality verified reports, and use them to patch the training environment.
Moonshot AI
A technical report on Kimi K3, a 2.8T-parameter sparse multimodal model with 104B active parameters and a one-million-token context window, post-trained with reinforcement learning across general, agentic, coding, and reasoning domains.
Jialian Li
A technical report on Athena-Brain-8B, an on-device embodied model trained through general supervised fine-tuning, general reinforcement learning, embodied-expert training, and model merging.
Z.ai
An official release of the 750B-A40B GLM-5.2 model, combining a one-million-token context, IndexShare sparse-attention reuse, and larger-scale agentic reinforcement learning for long-horizon tasks.
Ang Li
A model-family report on Ling-2.6 and Ring-2.6, combining architectural migration, long-context efficiency work, token-efficient reasoning objectives, and asynchronous agent reinforcement learning at trillion-parameter scale.
MiniMax
A technical report on the MiniMax-M2 family, pairing a 229.9B-parameter sparse MoE with agent-generated, verifiable trajectories and Forge, a scalable reinforcement-learning system for long-horizon agents.
Z.ai
An official GLM-5.1 model update focused on keeping an agent productive across longer coding and engineering runs through repeated execution, inspection, diagnosis, and strategy revision.
Cursor
A technical report on Composer 2, a specialized coding model trained through continued pretraining followed by large-scale reinforcement learning on long-horizon software-engineering tasks.
Z.ai
A GLM-5 technical report centered on agentic engineering, combining a more efficient long-context architecture with asynchronous reinforcement-learning infrastructure and agent RL for complex, long-horizon software tasks.
Tianbao Xie
OSWorld introduces a benchmark and environment for evaluating multimodal agents on open-ended tasks across real computer interfaces.
NVIDIA
Eureka uses a code-capable language model to propose and iteratively improve executable reward functions for reinforcement-learning tasks.
Princeton University
SWE-bench turns real GitHub issues and repository states into tasks for evaluating whether language models can produce working software patches.
Carnegie Mellon University
WebArena provides self-hosted, realistic websites and benchmark tasks for agents that must navigate interfaces, maintain state, and complete multi-step goals.
Guanzhi Wang
Voyager combines an automatic curriculum, an executable skill library, and iterative prompting to build an open-ended Minecraft agent without model parameter updates.
Google DeepMind
DreamerV3 presents a world-model-based reinforcement-learning algorithm designed to work across varied domains with a shared configuration.
NVIDIA
MineDojo combines a Minecraft simulation suite, a broad task set, and internet-scale multimodal knowledge for research on generalist embodied agents.
Michael Dennis
This work frames task-distribution design as an optimization problem and introduces PAIRED, where an environment generator uses regret to create solvable but challenging curricula.
OpenAI
The OpenAI Five report documents a distributed, continually trained self-play system for a long-horizon, imperfect-information team game.