# SpecReason: Fast and Accurate Inference-Time Compute via Speculative Reasoning

arXiv: https://arxiv.org/abs/2504.07891 · HTML: https://mlsanity.com/paper/2504.07891/

## Key facts

- Cited by 59 papers in the map (data through 14 Oct 2026).
- Placed in Test-time compute and efficient thinking, one of 17 directions in LLM reasoning.
- Builds on 24 of its 37 references that are in the map.
- First posted on arXiv 10 Apr 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)
- Efficient LLM inference and serving / [Speculative decoding](https://mlsanity.com/inference/speculative-decoding/)

## Builds on (most cited first, 24 of 24)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [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)
- [Fast Inference from Transformers via Speculative Decoding](https://mlsanity.com/paper/2211.17192/) (2022)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [Medusa: Simple LLM Inference Acceleration Framework with Multiple Decoding Heads](https://mlsanity.com/paper/2401.10774/) (2024)
- [SpecInfer: Accelerating Generative Large Language Model Serving with Tree-based Speculative Inference and Verification](https://mlsanity.com/paper/2305.09781/) (2023)
- [EAGLE-3: Scaling up Inference Acceleration of Large Language Models via Training-Time Test](https://mlsanity.com/paper/2503.01840/) (2025)
- [Blockwise Parallel Decoding for Deep Autoregressive Models](https://mlsanity.com/paper/1811.03115/) (2018)
- [rStar-Math: Small LLMs Can Master Math Reasoning with Self-Evolved Deep Thinking](https://mlsanity.com/paper/2501.04519/) (2025)
- [Break the Sequential Dependency of LLM Inference Using Lookahead Decoding](https://mlsanity.com/paper/2402.02057/) (2024)
- [REST: Retrieval-Based Speculative Decoding](https://mlsanity.com/paper/2311.08252/) (2023)
- [Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solvers](https://mlsanity.com/paper/2408.06195/) (2024)
- [PRMBench: A Fine-grained and Challenging Benchmark for Process-Level Reward Models](https://mlsanity.com/paper/2501.03124/) (2025)
- [Turning Trash into Treasure: Accelerating Inference of Large Language Models with Token Recycling](https://mlsanity.com/paper/2408.08696/) (2024)
- [Lookahead: An Inference Acceleration Framework for Large Language Model with Lossless Generation Accuracy](https://mlsanity.com/paper/2312.12728/) (2023)
- [Decoding Speculative Decoding](https://mlsanity.com/paper/2402.01528/) (2024)
- [Marconi: Prefix Caching for the Era of Hybrid LLMs](https://mlsanity.com/paper/2411.19379/) (2024)

## Built on by (most cited first, 30 of 59)

- [Efficient Reasoning Models: A Survey](https://mlsanity.com/paper/2504.10903/) (2025)
- [SelfBudgeter: Adaptive Token Allocation for Efficient LLM Reasoning](https://mlsanity.com/paper/2505.11274/) (2025)
- [Don't Overthink It: A Survey of Efficient R1-style Large Reasoning Models](https://mlsanity.com/paper/2508.02120/) (2025)
- [R2R: Efficiently Navigating Divergent Reasoning Paths with Small-Large Model Token Routing](https://mlsanity.com/paper/2505.21600/) (2025)
- [Agentic Plan Caching: Test-Time Memory for Fast and Cost-Efficient LLM Agents](https://mlsanity.com/paper/2506.14852/) (2025)
- [VeriThinker: Learning to Verify Makes Reasoning Model Efficient](https://mlsanity.com/paper/2505.17941/) (2025)
- [GlimpRouter: Efficient Collaborative Inference by Glimpsing One Token of Thoughts](https://mlsanity.com/paper/2601.05110/) (2026)
- [TrimR: Verifier-based Training-Free Thinking Compression for Efficient Test-Time Scaling](https://mlsanity.com/paper/2505.17155/) (2025)
- [Taming the Long-Tail: Efficient Reasoning RL Training with Adaptive Drafter](https://mlsanity.com/paper/2511.16665/) (2025)
- [AutoJudge: Judge Decoding Without Manual Annotation](https://mlsanity.com/paper/2504.20039/) (2025)
- [Scaling Speculative Decoding with Lookahead Reasoning](https://mlsanity.com/paper/2506.19830/) (2025)
- [Done Is Better than Perfect: Unlocking Efficient Reasoning by Structured Multi-Turn Decomposition](https://mlsanity.com/paper/2505.19788/) (2025)
- [Exploring the System 1 Thinking Capability of Large Reasoning Models](https://mlsanity.com/paper/2504.10368/) (2025)
- [Confidence-Guided Stepwise Model Routing for Cost-Efficient Reasoning](https://mlsanity.com/paper/2511.06190/) (2025)
- [Small Drafts, Big Verdict: Information-Intensive Visual Reasoning via Speculation](https://mlsanity.com/paper/2510.20812/) (2025)
- [RelayLLM: Efficient Reasoning via Collaborative Decoding](https://mlsanity.com/paper/2601.05167/) (2026)
- [TRIM: Hybrid Inference via Targeted Stepwise Routing in Multi-Step Reasoning Tasks](https://mlsanity.com/paper/2601.10245/) (2026)
- [Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models](https://mlsanity.com/paper/2601.11340/) (2026)
- [DualSpec: Accelerating Deep Research Agents via Dual-Process Action Speculation](https://mlsanity.com/paper/2603.07416/) (2026)
- [Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling](https://mlsanity.com/paper/2509.04474/) (2025)
- [Slipstream: Trajectory-Grounded Compaction Validation for Long-Horizon Agents](https://mlsanity.com/paper/2605.08580/) (2026)
- [Fast Quiet-STaR: Thinking Without Thought Tokens](https://mlsanity.com/paper/2505.17746/) (2025)
- [InfoSeeker: A Scalable Hierarchical Parallel Agent Framework for Web Information Seeking](https://mlsanity.com/paper/2604.02971/) (2026)
- [SSR: Speculative Parallel Scaling Reasoning in Test-time](https://mlsanity.com/paper/2505.15340/) (2025)
- [Efficient Tree-Structured Deep Research with Adaptive Resource Allocation](https://mlsanity.com/paper/2510.05145/) (2025)
- [Reject Only Critical Tokens: Pivot-Aware Speculative Decoding](https://mlsanity.com/paper/2511.00351/) (2025)
- [Fail Fast, Win Big: Rethinking the Drafting Strategy in Speculative Decoding via Diffusion LLMs](https://mlsanity.com/paper/2512.20573/) (2025)
- [ConfSpec: Efficient Step-Level Speculative Reasoning via Confidence-Gated Verification](https://mlsanity.com/paper/2602.18447/) (2026)
- [FastTTS: Accelerating Test-Time Scaling for Edge LLM Reasoning](https://mlsanity.com/paper/2509.00195/) (2025)
- [ATTS: Asynchronous Test-Time Scaling via Conformal Prediction](https://mlsanity.com/paper/2509.15148/) (2025)

## Abstract (arXiv)

Recent advances in inference-time compute have significantly improved performance on complex tasks by generating long chains of thought (CoTs) using Large Reasoning Models (LRMs). However, this improved accuracy comes at the cost of high inference latency due to the length of generated reasoning sequences and the autoregressive nature of decoding. Our key insight in tackling these overheads is that LRM inference, and the reasoning that it embeds, is highly tolerant of approximations: complex tasks are typically broken down into simpler steps, each of which brings utility based on the semantic insight it provides for downstream steps rather than the exact tokens it generates. Accordingly, we introduce SpecReason, a system that automatically accelerates LRM inference by using a lightweight model to (speculatively) carry out simpler intermediate reasoning steps and reserving the costly base model only to assess (and potentially correct) the speculated outputs. Importantly, SpecReason's focus on exploiting the semantic flexibility of thinking tokens in preserving final-answer accuracy is complementary to prior speculation techniques, most notably speculative decoding, which demands token-level equivalence at each step. Across a variety of reasoning benchmarks, SpecReason achieves 1.4-3.0× speedup over vanilla LRM inference while improving accuracy by 0.4-9.0%. Compared to speculative decoding without SpecReason, their combination yields an additional 8.8-58.0% latency reduction. We open-source SpecReason at https://github.com/ruipeterpan/specreason.