← Back to Research
arxiv
LLM February 22, 2026 Impact: 9/10

LIMO: Less is More for Reasoning

By Yixin Ye, Zhen Huang, Yang Xiao, Ethan Chern, Shijie Xia, Pengfei Liu
We challenge the prevailing assumption that complex reasoning in large language models (LLMs) necessitates massive training data. We demonstrate that sophisticated mathematical reasoning can emerge with only a few examples. Specifically, through simple supervised fine-tuning, our model, LIMO, achieves 63.3% accuracy on AIME24 and 95.6% on MATH500, surpassing previous fine-tuned models (6.5% on AIME24, 59.2% on MATH500) while using only 1% of the training data required by prior approaches. Furthermore, LIMO exhibits strong out-of-distribution generalization, achieving a 45.8% absolute improvement across diverse benchmarks, outperforming models trained on 100x more data. Synthesizing these findings, we propose the Less-Is-More Reasoning Hypothesis (LIMO Hypothesis): In foundation models where domain knowledge has been comprehensively encoded during pre-training, sophisticated reasoning can emerge through minimal but strategically designed demonstrations of cognitive processes.
reasoningfine-tuningmathematical reasoningdata efficiencyAIMECOLM 2025
Analysis

Summary

LIMO (Less Is More for Reasoning) is a striking challenge to one of the core assumptions driving modern AI training: that better reasoning requires more data. Researchers from Shanghai AI Lab demonstrate that a model fine-tuned on just 817 carefully selected examples can outperform models trained on datasets 100x larger on mathematical reasoning benchmarks. The key result — 63.3% on AIME24 versus the previous SFT SOTA of 6.5% — is the kind of jump that makes you do a double-take. This is not a marginal improvement; it's a category shift achieved with a fraction of the typical training budget.

Why It Matters

The LIMO Hypothesis proposed by the authors reframes the problem entirely: if a foundation model already has the relevant knowledge from pre-training, the job of fine-tuning is not to teach it new facts but to provide "cognitive templates" — demonstrations of how to reason through problems. This means the bottleneck isn't data volume but data quality and the selection strategy behind it. The paper's 45.8% absolute improvement in out-of-distribution generalization is equally compelling — it suggests LIMO isn't memorizing solution patterns but genuinely internalizing reasoning strategies that transfer.

Implications for AI Development

This paper has significant practical implications. If the LIMO Hypothesis holds broadly, it suggests we're in an era of pre-training abundance where foundation models carry enough latent knowledge that expensive large-scale SFT datasets may be unnecessary — or even counterproductive (adding noise). The field may shift toward curation and example quality over raw scale, with massive implications for training costs, data collection pipelines, and the role of human expert demonstration. It also reinforces the s1 and ReasonFlux findings: test-time and post-training efficiency improvements are unlocking performance gains that once required orders of magnitude more compute.

Critical Perspective

The critical question is generalizability beyond mathematical benchmarks. Math has clean ground-truth verification, making it ideal for studying reasoning quality — but whether the LIMO Hypothesis extends to open-ended reasoning, code generation, or multi-step planning remains to be demonstrated. The paper was accepted at COLM 2025, giving it solid peer-review credibility. Future work will likely probe how "cognitive templates" interact with different model sizes and pre-training corpora, and whether the 817-example sweet spot is universal or domain-specific.

Key Findings
1
63.3% on AIME24 with only 817 training examples — vs previous SFT SOTA of 6.5%
2
95.6% on MATH500 using 1% of the training data required by prior approaches
3
45.8% absolute improvement in out-of-distribution generalization over models trained on 100x more data
Impact Score
9/10
Expert Commentary
LIMO is one of the most important efficiency papers of 2025 — it reframes the role of fine-tuning from knowledge injection to cognitive template provision, suggesting that data curation quality will matter far more than data scale in the coming era of post-training optimization.
D
Dong Tran
AI Researcher. Deep-dive analysis of frontier AI research papers.