The Landscape of Agentic Reinforcement Learning for LLMs: A Survey
Guibin Zhang et al.
arXiv:2509.02547
In short
This survey argues agentic RL is a real shift from RL-tuning LLMs: the model acts over many steps in a partially observed world instead of producing one reply. It organises over 500 works by capability (planning, tools, memory, reasoning, self-improvement) and application, and catalogues environments and frameworks.
Why it matters
The map to read first if you want to understand how agents are trained, not just prompted.
Read first
The 4 Field Guide ideas this paper leans on.
Starting from scratch? The full route 26 ideas · basics first
- Machine Learning ✓ understood
Building systems that learn patterns from data instead of following hand-written rules, getting better at a task as they see more examples.
- Reinforcement Learning · read first ✓ understood
Learning through interaction with an environment, receiving rewards or penalties to learn optimal behavior policies.
- Reward ✓ understood
The single number an environment sends back after each action, telling a reinforcement learning agent how good that step was; the agent learns to maximize its long-run total.
- Agent ✓ understood
In reinforcement learning, the learner and decision-maker that observes an environment, takes actions and adjusts its behaviour to maximize cumulative reward.
- Environment ✓ understood
In RL, the world the agent interacts with, providing states, accepting actions, and returning rewards.
- Markov Decision Process · read first ✓ understood
A mathematical framework for modeling sequential decision-making with states, actions, rewards, and transition probabilities.
- Natural Language Processing ✓ understood
The field of AI that lets computers read, interpret, translate and generate human language, from spam filters and search to chatbots.
- Token ✓ understood
The unit of text a language model reads and writes: a word, part of a word, or a character, drawn from the model's fixed vocabulary.
- Tokenization ✓ understood
Splitting text into tokens, usually subword pieces, and mapping each to an integer ID so a language model can process it.
- Language Modeling ✓ understood
Assigning probabilities to sequences of text, in practice by predicting each next token from the ones before it; the objective behind large language models.
- Dataset ✓ understood
A collection of data examples used for training, validating, or testing machine learning models.
- Training ✓ understood
The process of fitting a model to data by repeatedly measuring how wrong its outputs are and adjusting its parameters to reduce that error.
- Unsupervised Learning ✓ understood
Learning from unlabeled data to discover hidden patterns, structures, or relationships without explicit target outputs.
- Self-Supervised Learning ✓ understood
Training on unlabeled data by making the data supply its own answers: hide or alter part of an input and have the model predict or match it.
- Pre-training ✓ understood
Training a model on a large dataset (often self-supervised) before fine-tuning on specific tasks, enabling transfer learning.
- Feature ✓ understood
A single measurable property of an example, such as a house's floor area or how many links an email contains, used as an input to a model.
- Neural Network ✓ understood
A model built from layers of simple units, each taking a weighted sum of its inputs and applying a nonlinearity; training adjusts the weights so the whole stack maps inputs to useful outputs.
- Deep Learning ✓ understood
A subset of machine learning that uses neural networks with multiple layers (deep neural networks) to learn hierarchical representations of data.
- Representation Learning ✓ understood
Letting a model learn its own internal description of the data, such as embeddings or layered visual features, instead of relying on features designed by hand.
- Embedding ✓ understood
A list of numbers (a vector) that represents a word, sentence, image or other item, learned so that similar items end up close together.
- Attention Mechanism ✓ understood
A technique that lets a neural network weigh every part of its input when producing each output, focusing on the parts most relevant at that step.
- Transformer ✓ understood
A neural network architecture, introduced in 2017, built from stacked self-attention and feed-forward layers; the basis of nearly every modern large language model.
- Large Language Model · read first ✓ understood
A neural network, almost always a transformer, trained on vast amounts of text to predict the next token, which lets it write, answer, summarize and follow instructions.
- Context Window ✓ understood
The maximum number of tokens a language model can take into account at once, counting both the input it reads and the output it writes. Also called context length.
- Tool Use ✓ understood
A language model asking the application around it to run a function, such as search, a calculator or an API, then using the result in its answer.
- AI Agent · read first ✓ understood
A system where a large language model decides its own next steps in a loop: calling tools, reading the results, and continuing until the task is done.
In the frontier
- Rank
- #21 of 100
- Citations
- 182
- as of Aug 9, 2026
- Published
- Sep 2025
Topics: RL for reasoning , Agent training and self-evolution , Agent benchmarks and computer use
Selection: 1kpapers.com by Together AI, most-cited as of Aug 9, 2026