Paper Detail

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

Natapong Nitarach

huggingface Score 5.5

Published 2026-04-16 · First seen 2026-04-17

General AI

Abstract

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.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@misc{nitarach2026model,
  title = {Model Capability Dominates: Inference-Time Optimization Lessons from AIMO 3},
  author = {Natapong Nitarach},
  year = {2026},
  abstract = {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},
  url = {https://huggingface.co/papers/2603.27844},
  keywords = {majority voting, mathematical reasoning, correlated errors, reasoning strategies, Diverse Prompt Mixer, AIMO 3 competition, high-temperature sampling, model capability, selection loss, verifier-based selector, code available, huggingface daily},
  eprint = {2603.27844},
  archiveprefix = {arXiv},
}

Metadata

{}