# Multiverse: Your Language Models Secretly Decide How to Parallelize and Merge Generation

arXiv: https://arxiv.org/abs/2506.09991 · HTML: https://mlsanity.com/paper/2506.09991/

## Key facts

- Cited by 30 papers in the map (data through 14 Oct 2026).
- Builds on 28 of its 47 references that are in the map.
- First posted on arXiv 11 Jun 2025.

## Builds on (most cited first, 28 of 28)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [Structured Denoising Diffusion Models in Discrete State-Spaces](https://mlsanity.com/paper/2107.03006/) (2021)
- [s1: Simple test-time scaling](https://mlsanity.com/paper/2501.19393/) (2025)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Large Language Diffusion Models](https://mlsanity.com/paper/2502.09992/) (2025)
- [Simple and Effective Masked Diffusion Language Models](https://mlsanity.com/paper/2406.07524/) (2024)
- [Discrete Diffusion Modeling by Estimating the Ratios of the Data Distribution](https://mlsanity.com/paper/2310.16834/) (2023)
- [Simplified and Generalized Masked Diffusion for Discrete Data](https://mlsanity.com/paper/2406.04329/) (2024)
- [LIMO: Less is More for Reasoning](https://mlsanity.com/paper/2502.03387/) (2025)
- [Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach](https://mlsanity.com/paper/2502.05171/) (2025)
- [Block Diffusion: Interpolating Between Autoregressive and Diffusion Language Models](https://mlsanity.com/paper/2503.09573/) (2025)
- [Remasking Discrete Diffusion Models with Inference-Time Scaling](https://mlsanity.com/paper/2503.00307/) (2025)
- [d1: Scaling Reasoning in Diffusion Large Language Models via Reinforcement Learning](https://mlsanity.com/paper/2504.12216/) (2025)
- [Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B](https://mlsanity.com/paper/2406.07394/) (2024)
- [Large Concept Models: Language Modeling in a Sentence Representation Space](https://mlsanity.com/paper/2412.08821/) (2024)
- [CLLMs: Consistency Large Language Models](https://mlsanity.com/paper/2403.00835/) (2024)
- [Learning Adaptive Parallel Reasoning with Language Models](https://mlsanity.com/paper/2504.15466/) (2025)
- [Theoretical Benefit and Limitation of Diffusion Language Model](https://mlsanity.com/paper/2502.09622/) (2025)
- [PENCIL: Long Thoughts with Short Memory](https://mlsanity.com/paper/2503.14337/) (2025)
- [Hogwild! Inference: Parallel LLM Generation via Concurrent Attention](https://mlsanity.com/paper/2504.06261/) (2025)
- [APE: Faster and Longer Context-Augmented Generation via Adaptive Parallel Encoding](https://mlsanity.com/paper/2502.05431/) (2025)
- [Learning to Keep a Promise: Scaling Language Model Decoding Parallelism with Learned Asynchronous Decoding](https://mlsanity.com/paper/2502.11517/) (2025)
- [Unifying Autoregressive and Diffusion-Based Sequence Generation](https://mlsanity.com/paper/2504.06416/) (2025)

## Built on by (most cited first, 30 of 30)

- [Parallel-R1: Towards Parallel Thinking via Reinforcement Learning](https://mlsanity.com/paper/2509.07980/) (2025)
- [Uni-cot: Towards Unified Chain-of-Thought Reasoning Across Text and Vision](https://mlsanity.com/paper/2508.05606/) (2025)
- [ParaThinker: Native Parallel Thinking as a New Paradigm to Scale LLM Test-time Compute](https://mlsanity.com/paper/2509.04475/) (2025)
- [Parallel-Probe: Towards Efficient Parallel Thinking via 2D Probing](https://mlsanity.com/paper/2602.03845/) (2026)
- [Native Parallel Reasoner: Reasoning in Parallelism via Self-Distilled Reinforcement Learning](https://mlsanity.com/paper/2512.07461/) (2025)
- [ASPD: Unlocking Adaptive Serial-Parallel Decoding by Exploring Intrinsic Parallelism in LLMs](https://mlsanity.com/paper/2508.08895/) (2025)
- [A Survey on Parallel Reasoning](https://mlsanity.com/paper/2510.12164/) (2025)
- [The Sequential Edge: Inverse-Entropy Voting Beats Parallel Self-Consistency at Matched Compute](https://mlsanity.com/paper/2511.02309/) (2025)
- [Planned Diffusion](https://mlsanity.com/paper/2510.18087/) (2025)
- [ThreadWeaver: Adaptive Threading for Efficient Parallel Reasoning in Language Models](https://mlsanity.com/paper/2512.07843/) (2025)
- [The Era of Agentic Organization: Learning to Organize with Language Models](https://mlsanity.com/paper/2510.26658/) (2025)
- [Generalized Parallel Scaling with Interdependent Generations](https://mlsanity.com/paper/2510.01143/) (2025)
- [Training Large Language Models To Reason In Parallel With Global Forking Tokens](https://mlsanity.com/paper/2510.05132/) (2025)
- [Accelerating Structured Chain-of-Thought in Autonomous Vehicles](https://mlsanity.com/paper/2602.02864/) (2026)
- [Multi-Stream LLMs: Unblocking Language Models with Parallel Streams of Thoughts, Inputs and Outputs](https://mlsanity.com/paper/2605.12460/) (2026)
- [OpenDeepThink: Parallel Reasoning via Bradley-Terry Aggregation](https://mlsanity.com/paper/2605.15177/) (2026)
- [Sharing State Between Prompts and Programs](https://mlsanity.com/paper/2512.14805/) (2025)
- [A2R: An Asymmetric Two-Stage Reasoning Framework for Parallel Reasoning](https://mlsanity.com/paper/2509.22044/) (2025)
- [On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency](https://mlsanity.com/paper/2601.21619/) (2026)
- [Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning](https://mlsanity.com/paper/2601.23027/) (2026)
- [MedVerse: Efficient and Reliable Medical Reasoning via DAG-Structured Parallel Execution](https://mlsanity.com/paper/2602.07529/) (2026)
- [One-Token Verification for Reasoning Correctness Estimation](https://mlsanity.com/paper/2603.01025/) (2026)
- [LaneRoPE: Positional Encoding for Collaborative Parallel Reasoning and Generation](https://mlsanity.com/paper/2605.27570/) (2026)
- [OPE: Overcoming Information Saturation in Parallel Thinking via Outline-Guided Path Exploration](https://mlsanity.com/paper/2602.08344/) (2026)
- [Visual Para-Thinker: Divide-and-Conquer Reasoning for Visual Comprehension](https://mlsanity.com/paper/2602.13310/) (2026)
- [Regulating Branch Parallelism in LLM Serving](https://mlsanity.com/paper/2605.06914/) (2026)
- [Second Thought: Reasoning in Parallel as LLM Agents Act and Observe](https://mlsanity.com/paper/2608.13667/) (2026)
- [Parason: Revealing Subtask and Trial Parallelism in LLM Reasoning](https://mlsanity.com/paper/2608.24658/) (2026)
- [Line-Coupled Language Model](https://mlsanity.com/paper/2609.07129/) (2026)
- [LLMs are General Asynchronous Agents](https://mlsanity.com/paper/2609.35427/) (2026)

## Abstract (arXiv)

Autoregressive Large Language Models (AR-LLMs) frequently exhibit implicit parallelism in sequential generation. Inspired by this, we introduce Multiverse, a new generative model that enables natively parallel generation. Multiverse internalizes a MapReduce paradigm, generating automatically through three stages: (i) a Map stage for adaptive task decomposition, (ii) a Process stage for parallel subtask execution, and (iii) a Reduce stage for lossless result synthesis. Next, we build a real-world Multiverse reasoning model with co-design of data, algorithm, and system, enabling rapid and seamless transfer from frontier AR-LLMs. For data creation, we develop Multiverse Curator, an automated LLM-assisted pipeline that transforms sequential reasoning chains into structured training data, avoiding costly human annotations. Algorithmically, we design Multiverse Attention to separate parallel reasoning steps while keeping compatibility with causal attention for efficient training. Systematically, we implement Multiverse Engine to support parallel inference. It features a dedicated interpreter that dynamically switches between sequential and parallel generation, triggered directly by the model. After a 3-hour fine-tuning with 1K examples, our Multiverse-32B stands as the only open-sourced non-AR model achieving performance on par with leading AR-LLMs of the same scale, evidenced by AIME24 & 25 scores of 54% and 46%, respectively. Moreover, our budget control experiments show that Multiverse-32B exhibits superior scaling, outperforming AR-LLMs by 1.87% on average using the same context length. Such scaling further leads to practical efficiency gains, achieving up to 2x speedup across varying batch sizes. We have open-sourced the entire Multiverse ecosystem, including data, model weights, engine, as well as complete data curation prompts and detailed training and evaluation recipes.