# An Empirical Study of LLM Reasoning Ability Under Strict Output Length Constraint

arXiv: https://arxiv.org/abs/2504.14350 · HTML: https://mlsanity.com/paper/2504.14350/

## Key facts

- Cited by 17 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 26 of its 49 references that are in the map.
- First posted on arXiv 19 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, 26 of 26)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Training Verifiers to Solve Math Word Problems](https://mlsanity.com/paper/2110.14168/) (2021)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [Qwen2 Technical Report](https://mlsanity.com/paper/2407.10671/) (2024)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone](https://mlsanity.com/paper/2404.14219/) (2024)
- [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://mlsanity.com/paper/2409.12122/) (2024)
- [Large Language Monkeys: Scaling Inference Compute with Repeated Sampling](https://mlsanity.com/paper/2407.21787/) (2024)
- [Reasoning with Language Model is Planning with World Model](https://mlsanity.com/paper/2305.14992/) (2023)
- [Phi-4 Technical Report](https://mlsanity.com/paper/2412.08905/) (2024)
- [Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs](https://mlsanity.com/paper/2412.21187/) (2024)
- [Chain of Draft: Thinking Faster by Writing Less](https://mlsanity.com/paper/2502.18600/) (2025)
- [Token-Budget-Aware LLM Reasoning](https://mlsanity.com/paper/2412.18547/) (2024)
- [AlphaMath Almost Zero: Process Supervision without Process](https://mlsanity.com/paper/2405.03553/) (2024)
- [Concise Thoughts: Impact of Output Length on LLM Reasoning and Cost](https://mlsanity.com/paper/2407.19825/) (2024)
- [T1: Advancing Language Model Reasoning through Reinforcement Learning and Inference Scaling](https://mlsanity.com/paper/2501.11651/) (2025)
- [Learning How Hard to Think: Input-Adaptive Allocation of LM Computation](https://mlsanity.com/paper/2410.04707/) (2024)
- [DriveCoT: Integrating Chain-of-Thought Reasoning with End-to-End Driving](https://mlsanity.com/paper/2403.16996/) (2024)
- [Adaptive Inference-Time Compute: LLMs Can Predict if They Can Do Better, Even Mid-Generation](https://mlsanity.com/paper/2410.02725/) (2024)
- [Make Every Penny Count: Difficulty-Adaptive Self-Consistency for Cost-Efficient Reasoning](https://mlsanity.com/paper/2408.13457/) (2024)
- [Evolving Deeper LLM Thinking](https://mlsanity.com/paper/2501.09891/) (2025)
- [Following Length Constraints in Instructions](https://mlsanity.com/paper/2406.17744/) (2024)
- [Precise Length Control in Large Language Models](https://mlsanity.com/paper/2412.11937/) (2024)

## Built on by (most cited first, 17 of 17)

- [ParaThinker: Native Parallel Thinking as a New Paradigm to Scale LLM Test-time Compute](https://mlsanity.com/paper/2509.04475/) (2025)
- [BudgetThinker: Empowering Budget-aware LLM Reasoning with Control Tokens](https://mlsanity.com/paper/2508.17196/) (2025)
- [VeriThinker: Learning to Verify Makes Reasoning Model Efficient](https://mlsanity.com/paper/2505.17941/) (2025)
- [Towards Concise and Adaptive Thinking in Large Reasoning Models: A Survey](https://mlsanity.com/paper/2507.09662/) (2025)
- [ThinkDial: An Open Recipe for Controlling Reasoning Effort in Large Language Models](https://mlsanity.com/paper/2508.18773/) (2025)
- [Controlling Thinking Speed in Reasoning Models](https://mlsanity.com/paper/2507.03704/) (2025)
- [Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking](https://mlsanity.com/paper/2509.23392/) (2025)
- [ORBIT: On-policy Exploration-Exploitation for Controllable Multi-Budget Reasoning](https://mlsanity.com/paper/2601.08310/) (2026)
- [Bridging the Knowledge-Prediction Gap in LLMs on Multiple-Choice Questions](https://mlsanity.com/paper/2509.23782/) (2025)
- [Measuring Weak-to-Strong Legibility of Reasoning Models](https://mlsanity.com/paper/2603.20508/) (2026)
- [MixReasoning: Switching Modes to Think](https://mlsanity.com/paper/2510.06052/) (2025)
- [Incorporating Self-Rewriting into Large Language Model Reasoning Reinforcement](https://mlsanity.com/paper/2511.16331/) (2025)
- [Mitigating LLM-based p-Hacking by Preregistering for the Next LLM](https://mlsanity.com/paper/2606.27687/) (2026)
- [MarkovScale: Towards Optimal Sequential Scaling at Inference Time](https://mlsanity.com/paper/2602.01120/) (2026)
- [Secure and Energy-Efficient Wireless Agentic AI Networks](https://mlsanity.com/paper/2602.15212/) (2026)
- [Keep, Customize, or Exit: Default Design and Token Pricing in LLM Reasoning Services](https://mlsanity.com/paper/2608.13315/) (2026)
- [RCCM: Diagnosing Gains and Harms in Reasoning-Prompt Transfer](https://mlsanity.com/paper/2610.19872/) (2026)

## Abstract (arXiv)

Recent work has demonstrated the remarkable potential of Large Language Models (LLMs) in test-time scaling. By making models think before answering, they are able to achieve much higher accuracy with extra inference computation. However, in many real-world scenarios, models are used under time constraints, where an answer should be given within a certain output length. It is unclear whether and how the reasoning ability of different LLMs remain effective under strict constraints. We take a first look at this problem by conducting an in-depth empirical study. Specifically, we test 30 LLMs on common reasoning datasets under a wide range of output length budgets, and we analyze the correlation between the inference accuracy and various properties including model type, model size, prompt style, etc. We also consider the mappings between token budgets and actual on-device latency budgets. The results have demonstrated several interesting findings regarding the budget-aware LLM reasoning ability that differ from the unconstrained situation, e.g. the optimal choices of either model size or prompt style change under different budgets. These findings offer timely evaluation to this area and practical guidance for users to deploy LLMs under real-world latency constraints.