# Efficiency Hallucination: Formalizing and Measuring Behavioral Calibration in LLM-Based Code Optimization

arXiv: https://arxiv.org/abs/2609.14839 · HTML: https://mlsanity.com/paper/2609.14839/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 14 of its 20 references that are in the map.
- First posted on arXiv 13 Sep 2026.

## Builds on (most cited first, 14 of 14)

- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Why Language Models Hallucinate](https://mlsanity.com/paper/2509.04664/) (2025)
- [Learning Performance-Improving Code Edits](https://mlsanity.com/paper/2302.07867/) (2023)
- [LLM Hallucinations in Practical Code Generation: Phenomena, Mechanism, and Mitigation](https://mlsanity.com/paper/2409.20550/) (2024)
- [EffiBench: Benchmarking the Efficiency of Automatically Generated Code](https://mlsanity.com/paper/2402.02037/) (2024)
- [LLaMoCo: Instruction Tuning of Large Language Models for Optimization Code Generation](https://mlsanity.com/paper/2403.01131/) (2024)
- [How Efficient is LLM-Generated Code? A Rigorous & High-Standard Benchmark](https://mlsanity.com/paper/2406.06647/) (2024)
- [SWE-Perf: Can Language Models Optimize Code Performance on Real-World Repositories?](https://mlsanity.com/paper/2507.12415/) (2025)
- [CodeMirage: Hallucinations in Code Generated by Large Language Models](https://mlsanity.com/paper/2408.08333/) (2024)
- [Efficient Hallucination Detection for LLMs Using Uncertainty-Aware Attention Heads](https://mlsanity.com/paper/2505.20045/) (2025)
- [Controlled Self-Evolution for Algorithmic Code Optimization](https://mlsanity.com/paper/2601.07348/) (2026)
- [PerfRL: A Small Language Model Framework for Efficient Code Optimization](https://mlsanity.com/paper/2312.05657/) (2023)
- [Evolving Excellence: Automated Optimization of LLM-based Agents](https://mlsanity.com/paper/2512.09108/) (2025)
- [How Do Agents Perform Code Optimization? An Empirical Study](https://mlsanity.com/paper/2512.21757/) (2025)

## Abstract (arXiv)

The integration of Large Language Models (LLMs) into automated code optimization introduces a critical reliability risk we term the Efficiency Hallucination: an LLM's tendency to issue non-functional mutations with unsubstantiated performance claims on already-optimized code. This is driven by the Evaluation Trap, wherein binary benchmarks incentivize unnecessary modifications over safely abstaining. We present a validation framework using classification penalty methods, evaluated across 180 optimization runs on nine models (GPT, Claude, Gemini) using EffiBench. Under standard prompts, models exhibit a 100% over-edit rate on optimal code. Our guardrail raises correct abstention from 0% to to 44.4%, preserving a 100% edit rate on sub-optimal code with zero false abstentions. Calibration is uneven: GPT-5.4 Mini approaches near-perfect abstention, and simple code is recognized more reliably than complex code. Our framework offers a training-free mechanism to mitigate LLM overconfidence before deployment in production.