# Rethinking Inference-Time Scaling: Efficiency Limits and Linguistic Signals

arXiv: https://arxiv.org/abs/2504.14047 · HTML: https://mlsanity.com/paper/2504.14047/

## Key facts

- Cited by 18 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 21 of its 34 references that are in the map.
- First posted on arXiv 18 Apr 2025.

## Placed in

- LLM reasoning / [Test-time compute and efficient thinking](https://mlsanity.com/lr/test-time-compute-and-efficient-thinking/)

## Builds on (most cited first, 21 of 21)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [LLaMA: Open and Efficient Foundation Language Models](https://mlsanity.com/paper/2302.13971/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [Qwen Technical Report](https://mlsanity.com/paper/2309.16609/) (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)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [OpenAI o1 System Card](https://mlsanity.com/paper/2412.16720/) (2024)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Gemini: A Family of Highly Capable Multimodal Models](https://mlsanity.com/paper/2312.11805/) (2023)
- [DSPy: Compiling Declarative Language Model Calls into Self-Improving Pipelines](https://mlsanity.com/paper/2310.03714/) (2023)
- [Mixture-of-Agents Enhances Large Language Model Capabilities](https://mlsanity.com/paper/2406.04692/) (2024)
- [Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning](https://mlsanity.com/paper/2502.14768/) (2025)
- [Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling](https://mlsanity.com/paper/2502.06703/) (2025)
- [Branch-Solve-Merge Improves Large Language Model Evaluation and Generation](https://mlsanity.com/paper/2310.15123/) (2023)
- [Prover-Verifier Games improve legibility of LLM outputs](https://mlsanity.com/paper/2407.13692/) (2024)
- [Planning In Natural Language Improves LLM Search For Code Generation](https://mlsanity.com/paper/2409.03733/) (2024)

## Built on by (most cited first, 18 of 18)

- [Don't Overthink it. Preferring Shorter Thinking Chains for Improved LLM Reasoning](https://mlsanity.com/paper/2505.17813/) (2025)
- [Inference-Time Hyper-Scaling with KV Cache Compression](https://mlsanity.com/paper/2506.05345/) (2025)
- [VeriThinker: Learning to Verify Makes Reasoning Model Efficient](https://mlsanity.com/paper/2505.17941/) (2025)
- [Interleaved Reasoning for Large Language Models via Reinforcement Learning](https://mlsanity.com/paper/2505.19640/) (2025)
- [Value-Guided Search for Efficient Chain-of-Thought Reasoning](https://mlsanity.com/paper/2505.17373/) (2025)
- [How Much Backtracking is Enough? Exploring the Interplay of SFT and RL in Enhancing LLM Reasoning](https://mlsanity.com/paper/2505.24273/) (2025)
- [When Life Gives You Samples: The Benefits of Scaling up Inference Compute for Multilingual LLMs](https://mlsanity.com/paper/2506.20544/) (2025)
- [LSPO: Length-aware Dynamic Sampling for Policy Optimization in LLM Reasoning](https://mlsanity.com/paper/2510.01459/) (2025)
- [Strategic Scaling of Test-Time Compute: A Bandit Learning Approach](https://mlsanity.com/paper/2506.12721/) (2025)
- [Neural Chain-of-Thought Search: Searching the Optimal Reasoning Path to Enhance Large Language Models](https://mlsanity.com/paper/2601.11340/) (2026)
- [DynScaling: Efficient Verifier-free Inference Scaling via Dynamic and Integrated Sampling](https://mlsanity.com/paper/2506.16043/) (2025)
- [From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI](https://mlsanity.com/paper/2606.14502/) (2026)
- [When to Ensemble: Identifying Token-Level Points for Stable and Fast LLM Ensembling](https://mlsanity.com/paper/2510.15346/) (2025)
- [Rethinking the Unsolvable: When In-Context Search Meets Test-Time Scaling](https://mlsanity.com/paper/2505.22290/) (2025)
- [Semi-structured LLM Reasoners Can Be Rigorously Audited](https://mlsanity.com/paper/2505.24217/) (2025)
- [Adaptive Test-Time Compute Allocation with Evolving In-Context Demonstrations](https://mlsanity.com/paper/2604.21018/) (2026)
- [MarkovScale: Towards Optimal Sequential Scaling at Inference Time](https://mlsanity.com/paper/2602.01120/) (2026)
- [CIPHER: A Decoupled Exploration-Selection Framework for Test-Time Scaling of Data Science Agents](https://mlsanity.com/paper/2607.14386/) (2026)

## Abstract (arXiv)

There is intense interest in investigating how inference time compute (ITC) (e.g. repeated sampling, refinements, etc) can improve large language model (LLM) capabilities. While breakthroughs like DeepSeek-R1 highlight the power of reinforcement learning for reasoning, the interaction between ITC and reasoning-optimized weights remains poorly understood. This work conducts a comprehensive analysis of inference-time scaling methods for both reasoning and non-reasoning models on challenging reasoning tasks. While prior work suggests that scaling test-time compute can optimally substitute for model parameter scaling, we identify a fundamental limit to this compute-equivalence, the reasoning floor, a performance plateau that non-reasoning models cannot escape, no matter how much inference compute is spent. We demonstrate that general-purpose models fail to match the accuracy of reasoning-optimized models even with an order of magnitude more inference compute, suggesting that internalizing reasoning protocols is a prerequisite for effective test-time scaling. Within reasoning models, we find that the complexity of the scaling method often yields diminishing returns; simple majority voting consistently outperforms sophisticated sequential revision and mixture-of-agents frameworks. Crucially, we identify a Linguistic Signal of Correctness - correct responses are significantly more concise and exhibit a lower density of hedging and thinking markers. We demonstrate that these intrinsic linguistic features can serve as zero-compute proxies for response quality, providing a pathway to more efficient, self-diagnostic reasoning agents.