Topic index

The learning loop,
dimension by dimension.

Eight durable lenses for understanding how reinforcement learning systems improve, plateau, and fail as they scale.

01

RL Scaling

RL Scaling studies what happens when reinforcement learning systems grow across compute, data, model capacity, rollout volume, task diversity, and feedback quality. The central question is not simply whether more resources improve a policy, but which part of the learning loop becomes the next constraint. This topic tracks empirical scaling behavior, large training systems, and the operational lessons behind durable gains.

35 items
Explore →
02

RLVR

Reinforcement Learning with Verifiable Rewards uses outcomes that can be checked by programs, tests, simulators, or formal rules. Math, code, and tool-use tasks are common examples. RLVR can make feedback cheaper and more scalable than human preference labeling, but only where the verifier captures the behavior that matters and resists reward gaming.

15 items
Explore →
03

GRPO & Policy Optimization

Group Relative Policy Optimization and related methods update a language-model policy using rewards from sampled response groups rather than a separately learned value critic. The family is attractive for reasoning post-training because it can simplify the RL stack, but performance still depends on sampling, normalization, clipping, KL control, and the quality of the reward signal.

11 items
Explore →
04

Agentic RL

Agentic RL applies reinforcement learning to systems that plan, use tools, interact with software, and act over long horizons. The hard problems extend beyond the base model: state, memory, environment resets, credit assignment, reliable verifiers, and safe exploration all become part of the training system.

20 items
Explore →
05

Environments

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.

19 items
Explore →
06

Rewards & Verifiers

Rewards translate outcomes and preferences into the signal used to improve a policy. They may come from humans, learned reward models, AI judges, tests, or executable verifiers. This topic follows how reward signals scale, where they fail, and how evaluation can detect proxy optimization before it becomes a production problem.

15 items
Explore →
07

Post-training

Post-training turns a pretrained model into a useful system through supervised fine-tuning, preference optimization, reinforcement learning, distillation, and targeted data generation. The choice of method should follow the desired behavior, available feedback, and evaluation setup rather than a preference for one algorithm.

25 items
Explore →
08

Scaling Laws

Scaling laws describe empirical relationships between resources and performance. In reinforcement learning, useful laws must account for interaction data, reward quality, environment diversity, optimization pressure, and failure under proxy objectives—not only parameter count and training compute.

14 items
Explore →