# Self-Evolving Deep Research via Joint Generation and Evaluation

arXiv: https://arxiv.org/abs/2606.04507 · HTML: https://mlsanity.com/paper/2606.04507/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Self-rewarding alignment and self-training, one of 13 directions in Synthetic data and self-improvement for LLMs.
- Builds on 39 of its 68 references that are in the map.
- First posted on arXiv 3 Jun 2026.

## Placed in

- Synthetic data and self-improvement for LLMs / [Self-rewarding alignment and self-training](https://mlsanity.com/synth/self-rewarding-alignment-and-self-training/)

## Builds on (most cited first, 30 of 39)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning](https://mlsanity.com/paper/2503.09516/) (2025)
- [Prometheus: Inducing Fine-grained Evaluation Capability in Language Models](https://mlsanity.com/paper/2310.08491/) (2023)
- [Search-o1: Agentic Search-Enhanced Large Reasoning Models](https://mlsanity.com/paper/2501.05366/) (2025)
- [Absolute Zero: Reinforced Self-play Reasoning with Zero Data](https://mlsanity.com/paper/2505.03335/) (2025)
- [WebThinker: Empowering Large Reasoning Models with Deep Research Capability](https://mlsanity.com/paper/2504.21776/) (2025)
- [R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2503.05592/) (2025)
- [Meta-Harness: End-to-End Optimization of Model Harnesses](https://mlsanity.com/paper/2603.28052/) (2026)
- [DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environments](https://mlsanity.com/paper/2504.03160/) (2025)
- [DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents](https://mlsanity.com/paper/2506.11763/) (2025)
- [R-Zero: Self-Evolving Reasoning LLM from Zero Data](https://mlsanity.com/paper/2508.05004/) (2025)
- [Learning to Reason without External Rewards](https://mlsanity.com/paper/2505.19590/) (2025)
- [Seedance 2.0: Advancing Video Generation for World Complexity](https://mlsanity.com/paper/2604.14148/) (2026)
- [Assisting in Writing Wikipedia-like Articles From Scratch with Large Language Models](https://mlsanity.com/paper/2402.14207/) (2024)
- [A Survey on Code Generation with LLM-based Agents](https://mlsanity.com/paper/2508.00083/) (2025)
- [SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learning](https://mlsanity.com/paper/2506.24119/) (2025)
- [The Dawn After the Dark: An Empirical Study on Factuality Hallucination in Large Language Models](https://mlsanity.com/paper/2401.03205/) (2024)
- [SimpleDeepSearcher: Deep Information Seeking via Web-Powered Reasoning Trajectory Synthesis](https://mlsanity.com/paper/2505.16834/) (2025)
- [SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Data](https://mlsanity.com/paper/2505.20347/) (2025)
- [Dr. Zero: Self-Evolving Search Agents without Training Data](https://mlsanity.com/paper/2601.07055/) (2026)
- [EvolveSearch: An Iterative Self-Evolving Search Agent](https://mlsanity.com/paper/2505.22501/) (2025)
- [Into the Unknown Unknowns: Engaged Human Learning through Participation in Language Model Agent Conversations](https://mlsanity.com/paper/2408.15232/) (2024)
- [Search Self-play: Pushing the Frontier of Agent Capability without Supervision](https://mlsanity.com/paper/2510.18821/) (2025)
- [On the Effect of Negative Gradient in Group Relative Deep Reinforcement Optimization](https://mlsanity.com/paper/2505.18830/) (2025)
- [How Far Can Unsupervised RLVR Scale LLM Training?](https://mlsanity.com/paper/2603.08660/) (2026)
- [Beyond Verifiable Rewards: Scaling Reinforcement Learning for Language Models to Unverifiable Data](https://mlsanity.com/paper/2503.19618/) (2025)
- [A Survey on Self-play Methods in Reinforcement Learning](https://mlsanity.com/paper/2408.01072/) (2024)
- [3DSceneEditor: Controllable 3D Scene Editing with Gaussian Splatting](https://mlsanity.com/paper/2412.01583/) (2024)
- [SPaR: Self-Play with Tree-Search Refinement to Improve Instruction-Following in Large Language Models](https://mlsanity.com/paper/2412.11605/) (2024)

## Built on by (most cited first, 2 of 2)

- [Janus: An Algorithm-Evaluator Co-Evolution Framework for LLM-Driven Discovery under Expensive Evaluation Budgets](https://mlsanity.com/paper/2608.08189/) (2026)
- [Mint-Agent: Introducing Finance-Native Agentic Foundation Models](https://mlsanity.com/paper/2608.16386/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have become increasingly adopted in daily applications, with deep research standing out as a particularly important capability. Unlike traditional question-answering (QA) tasks, deep research report generation lacks definitive ground-truth, making reward design inherently unverifiable and limiting effective reinforcement learning. Existing approaches mitigate this challenge with LLM-as-a-judge and query-dependent evaluation rubrics, but they still rely on static evaluators that cannot adapt their standards as the solver improves, leading to insufficient and eventually saturated optimization pressure. We address this limitation with a self-evolving co-evolutionary training framework for deep research evaluation and generation (SCORE), which tightly couples an evaluator and a solver in a shared-parameter learning process. Rather than treating generation and evaluation as isolated modules, we leverage their intrinsic connection to enable joint improvement within a single shared-parameter model. To restrict this process, we introduce a meta-harness, which dynamically controls the evaluation environment based on solver performance, encouraging valid evaluation dimensions and sufficiently deep evaluator search. Extensive experiments on deep research benchmarks demonstrate consistent improvement in report generation quality, showing that co-evolving evaluation and generation is a promising direction for training open-ended research agents.