mlsanityData to 7 Oct 2026

Loong: Synthesize Long Chain-of-Thoughts at Scale through Verifiers

arXiv 2509.03059 ↗First posted 3 Sep 2025cs.LG cs.AIcited by 14

On the map: LLM reasoning → Reinforcement learning for reasoning

Abstract (arXiv)

Recent advances in Large Language Models (LLMs) have shown that their reasoning capabilities can be significantly improved through Reinforcement Learning with Verifiable Reward (RLVR), particularly in domains like mathematics and programming, where ground-truth correctness can be automatically evaluated. However, extending this success to other reasoning-intensive domains remains challenging due to the scarcity of high-quality, verifiable datasets and the high cost of human supervision. In this work, we introduce the Loong Project: an open-source framework for scalable synthetic data generation and verification across a diverse range of reasoning-intensive domains. The framework consists of two key components: (1) LoongBench, a curated seed dataset containing 8,729 human-vetted examples across 12 domains (e.g., Advanced Mathematics, Chemistry, Logic), each paired with executable code and rich metadata; and (2) LoongEnv, a modular synthetic data generation environment that supports multiple prompting strategies to produce new question-answer-code triples. Together, these components form an agent-environment loop that enables reinforcement learning, where an LLM-based agent is rewarded for generating Chain-of-Thought (CoT) solutions that align with code-executed answers. Empirically, we benchmark LoongBench on a broad suite of both open-source and proprietary LLMs to evaluate domain coverage and reveal performance bottlenecks. In addition, we conduct a comprehensive analysis of synthetic data generated by LoongEnv, examining correctness, difficulty, and diversity. Code and documentation are available at https://github.com/camel-ai/loong.

Key facts. Cited by 14 papers in the map (data through 7 Oct 2026). Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning. Builds on 23 of its 60 references that are in the map. First posted on arXiv 3 Sep 2025.

Builds on

23 of 60 references.

Closest ancestors · 7

Same direction, newest first

  1. 2025ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models30 May
  2. 2025Reinforcement Learning for Reasoning in Large Language Models with One Training Example29 Apr
  3. 2025Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains31 Mar
  4. 2025Reinforcement Learning with Verifiable Rewards: GRPO's Effective Loss, Dynamics, and Success Amplification9 Mar
  5. 2025Satori: Reinforcement Learning with Chain-of-Action-Thought Enhances LLM Reasoning via Autoregressive Search4 Feb
  6. 2025DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning22 Jan
  7. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Model29 May

Other references · 16

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  2. 2025Qwen3 Technical Reportcited by 8,195
  3. 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,674
  4. 2024A Survey on LLM-as-a-Judgecited by 1,384
  5. 2023WizardLM: Empowering large pre-trained language models to follow complex instructionscited by 1,116
  6. 2023LIMA: Less Is More for Alignmentcited by 1,080
  7. 2023The Flan Collection: Designing Data and Methods for Effective Instruction Tuningcited by 734
  8. 2023WizardCoder: Empowering Code Large Language Models with Evol-Instructcited by 728
  9. 2023RRHF: Rank Responses to Align Language Models with Human Feedback without tearscited by 454
  10. 2023Preference Ranking Optimization for Human Alignmentcited by 283
  11. 2024Agent-Pro: Learning to Evolve via Policy-Level Reflection and Optimizationcited by 93
  12. 2025LeetCodeDataset: A Temporal Dataset for Robust Evaluation and Efficient Training of Code LLMscited by 67
4 more
  1. 2024MedCalc-Bench: Evaluating Large Language Models for Medical Calculationscited by 64
  2. 2025Synthetic Data Generation Using Large Language Models: Advances in Text and Codecited by 61
  3. 2025Co-Evolving LLM Coder and Unit Tester via Reinforcement Learningcited by 51
  4. 2021AR-LSAT: Investigating Analytical Reasoning of Textcited by 49

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2025: 2; 2026: 12.

14 papers.

Sort:

By area:

  1. 2025Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learningcited by 16
  2. 2026MetaSkill-Evolve: Recursive Self-Improvement of LLM Agents via Two-Timescale Meta-Skill Evolutioncited by 12
  3. 2026OPD-V: Visual On-Policy Self-Distillation with Modality Balancecited by 7
  4. 2025SafeRBench: Dissecting the Reasoning Safety of Large Language Modelscited by 4
  5. 2026EchoRL: Reinforcement Learning via Rollout Echoingcited by 3
  6. 2026Reasoning on the Manifold: Bidirectional Consistency for Self-Verification in Diffusion Language Modelscited by 2
  7. 2026ReflectRL: Learning from Golden Negative Trajectories via Reflective-to-Direct Reasoningcited by 2
  8. 2026MemWM: Memory-Augmented Text-Based World Modelcited by 2
  9. 2026Can Large Language Models Generalize Procedures Across Representations?cited by 1
  10. 2026Uncertainty-Aware Abstention in Large Language Models with Provable Alignment Guaranteescited by 1
  11. 2026Sci-CoE: Co-evolving Scientific Reasoning LLMs via Geometric Consensus with Sparse Supervisioncited by 0
  12. 2026DocCogito: Aligning Layout Cognition and Step-Level Grounded Reasoning for Document Understandingcited by 0
2 more
  1. 2026SeDa: A Unified System for Dataset Discovery and Multi-Entity Augmented Semantic Explorationcited by 0
  2. 2026Asymptotic Risk Calibration for Selective Question Answeringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025LoongRL: Reinforcement Learning for Advanced Reasoning over Long Contexts
  2. 2025VerifyBench: A Systematic Benchmark for Evaluating Reasoning Verifiers Across Domains
  3. 2026Learning from Synthetic Data Improves Multi-hop Reasoning
  4. 2025Enigmata: Scaling Logical Reasoning in Large Language Models with Synthetic Verifiable Puzzles
  5. 2025EvoSyn: Generalizable Evolutionary Data Synthesis for Verifiable Learning
  6. 2025Reinforcing General Reasoning without Verifiers
  7. 2026Re2: Unlocking LLM Reasoning via Reinforcement Learning with Re-solving
  8. 2026ReSyn: Autonomously Scaling Synthetic Environments for Reasoning Models