Agents & RL Nov 2025 · #73 most cited · 75 citations

DR Tulu: Reinforcement Learning with Evolving Rubrics for Deep Research

Rulin Shao et al.

arXiv:2511.19399

In short

Long-form research answers can’t be checked against a single right answer, so DR Tulu is trained with rubrics that evolve alongside the model, drawing on what it finds while searching. The resulting 8B open model matches proprietary deep-research agents at a tiny fraction of the cost per query.

Why it matters

It shows how to use RL where there is no verifiable answer.

Read first

The 3 Field Guide ideas this paper leans on.

Starting from scratch? The full route 23 ideas · basics first
  1. 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.

  2. Reinforcement Learning · read first ✓ understood

    Learning through interaction with an environment, receiving rewards or penalties to learn optimal behavior policies.

  3. Reward · read first ✓ 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.

  4. 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.

  5. 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.

  6. Tokenization ✓ understood

    Splitting text into tokens, usually subword pieces, and mapping each to an integer ID so a language model can process it.

  7. 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.

  8. Dataset ✓ understood

    A collection of data examples used for training, validating, or testing machine learning models.

  9. 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.

  10. Unsupervised Learning ✓ understood

    Learning from unlabeled data to discover hidden patterns, structures, or relationships without explicit target outputs.

  11. 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.

  12. Pre-training ✓ understood

    Training a model on a large dataset (often self-supervised) before fine-tuning on specific tasks, enabling transfer learning.

  13. 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.

  14. 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.

  15. Deep Learning ✓ understood

    A subset of machine learning that uses neural networks with multiple layers (deep neural networks) to learn hierarchical representations of data.

  16. 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.

  17. 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.

  18. 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.

  19. 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.

  20. Large Language Model ✓ 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.

  21. 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.

  22. 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.

  23. 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
#73 of 100
Citations
75
as of Aug 9, 2026
Published
Nov 2025

Topics: Search and deep research , RL for reasoning , Agent training and self-evolution

Selection: 1kpapers.com by Together AI, most-cited as of Aug 9, 2026

Nearby papers

Summary in our own words; read the paper for the details. ← All papers