# Towards Evaluation of Implicit Software World Models in Coding LLMs

arXiv: https://arxiv.org/abs/2606.27406 · HTML: https://mlsanity.com/paper/2606.27406/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in World models for language agents, one of 11 directions in World models.
- Builds on 9 of its 23 references that are in the map.
- First posted on arXiv 25 Jun 2026.

## Placed in

- World models / [World models for language agents](https://mlsanity.com/wm/world-models-for-language-agents/)

## Builds on (most cited first, 9 of 9)

- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [CWM: An Open-Weights LLM for Research on Code Generation with World Models](https://mlsanity.com/paper/2510.02387/) (2025)
- [Commit0: Library Generation from Scratch](https://mlsanity.com/paper/2412.01769/) (2024)
- [CodeMind: Evaluating Large Language Models for Code Reasoning](https://mlsanity.com/paper/2402.09664/) (2024)
- [CodeCrash: Exposing LLM Fragility to Misleading Natural Language in Code Reasoning](https://mlsanity.com/paper/2504.14119/) (2025)
- [BigO(Bench): Can LLMs Generate Code with Controlled Time and Space Complexity?](https://mlsanity.com/paper/2503.15242/) (2025)
- [Evaluating Code Reasoning Abilities of Large Language Models Under Real-World Settings](https://mlsanity.com/paper/2512.14917/) (2025)
- [ThrowBench: Benchmarking LLMs by Predicting Runtime Exceptions](https://mlsanity.com/paper/2503.04241/) (2025)

## Abstract (arXiv)

Software engineering, whether performed by humans or by AI agents, requires reasoning about how software behaves. We call the internal model that supports such reasoning the software world model, and view current code-execution benchmarks as covering one well-studied slice of it -- control flow. In this paper, we take a step toward a broader evaluation by shifting the observable axis to execution resources: alongside test outcome and exception class, we predict peak memory, wall-clock time, and ranked profiler outputs at method and line granularity. We use SWE-bench Verified as the source of data to hold the test close to real-world software engineering tasks. All tested models, frontier ones included, show modest performance and brittle behaviour, suggesting a notable lack of understanding of how software is executed, as opposed to how its source code is written.