mlsanityData to 14 Oct 2026

Model Capability Dominates: Inference-Time Optimization Lessons from AIMO 3

arXiv 2603.27844 ↗First posted 29 Mar 2026cs.CLcited by 2

On the map: LLM reasoning → Test-time compute and efficient thinking

Abstract (arXiv)

Majority voting over multiple LLM attempts improves mathematical reasoning, but correlated errors limit the effective sample size. A natural fix is to assign different reasoning strategies to different voters. The approach, Diverse Prompt Mixer, is tested on the AIMO 3 competition: 3 models, 23+ experiments, 50 IMO-level problems, one H100 80 GB, 5-hour limit. Every prompt-level intervention fails. High-temperature sampling already decorrelates errors; weaker strategies reduce accuracy more than they reduce correlation. Across an 8-point capability gap at equal N=8 and every optimization tested, model capability dominates. The gap between the best majority-vote score (42/50) and pass@20 (~45.5) is selection loss, not prompt loss. A verifier-based selector could close it. Prompt engineering cannot.

Key facts. Cited by 2 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 8 of its 24 references that are in the map. First posted on arXiv 29 Mar 2026.

Builds on

8 of 24 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters6 Aug

Other references · 7

  1. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,944
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  3. 2024Large Language Monkeys: Scaling Inference Compute with Repeated Samplingcited by 991
  4. 2024Gated Delta Networks: Improving Mamba2 with Delta Rulecited by 532
  5. 2024OpenMathInstruct-2: Accelerating AI for Math with Massive Open-Source Instruction Datacited by 200
  6. 2025AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning datasetcited by 174
  7. 2025Towards LLMs Robustness to Changes in Prompt Format Stylescited by 21

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026When More Sampling Hurts: The Modal Ceiling and Correlation Ceiling of Test-Time Scalingcited by 6
  2. 2026The Exceedance Design Effect: Effective Sample Size for Thresholds under Clusteringcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Rethinking the Role of Prompting Strategies in LLM Test-Time Scaling: A Perspective of Probability Theory
  2. 2026The Capability Frontier: Benchmarks Miss 82% of Model Performance
  3. 2025Mitigating Strategy-Selection Bias in Reasoning for More Effective Test-Time Scaling
  4. 2025On the Effect of Sampling Diversity in Scaling LLM Inference
  5. 2026Ranking Reasoning LLMs under Test-Time Scaling
  6. 2025Best-of-Majority: Minimax-Optimal Strategy for Pass@k Inference Scaling
  7. 2025Can 1B LLM Surpass 405B LLM? Rethinking Compute-Optimal Test-Time Scaling
  8. 2026AI4AI at Test-Time: Strong-to-Weak Capability Transfer via Harnesses