# Are LLMs Ready for Neural-integrated Mechanistic Modeling? A Benchmark and Agentic Framework

arXiv: https://arxiv.org/abs/2602.18008 · HTML: https://mlsanity.com/paper/2602.18008/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 17 of its 49 references that are in the map.
- First posted on arXiv 20 Feb 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 17 of 17)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Efficient Memory Management for Large Language Model Serving with PagedAttention](https://mlsanity.com/paper/2309.06180/) (2023)
- [Tree of Thoughts: Deliberate Problem Solving with Large Language Models](https://mlsanity.com/paper/2305.10601/) (2023)
- [AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://mlsanity.com/paper/2506.13131/) (2025)
- [Estimating Counterfactual Treatment Outcomes over Time Through Adversarially Balanced Representations](https://mlsanity.com/paper/2002.04083/) (2020)
- [Causal Transformer for Estimating Counterfactual Outcomes](https://mlsanity.com/paper/2204.07258/) (2022)
- [Continuous-Time Modeling of Counterfactual Outcomes Using Neural Controlled Differential Equations](https://mlsanity.com/paper/2206.08311/) (2022)
- [BioCoder: A Benchmark for Bioinformatics Code Generation with Large Language Models](https://mlsanity.com/paper/2308.16458/) (2023)
- [Differentiable Agent-based Epidemiology](https://mlsanity.com/paper/2207.09714/) (2022)
- [CAMul: Calibrated and Accurate Multi-view Time-Series Forecasting](https://mlsanity.com/paper/2109.07438/) (2021)
- [Automatically Learning Hybrid Digital Twins of Dynamical Systems](https://mlsanity.com/paper/2410.23691/) (2024)
- [High-resolution synthetic residential energy use profiles for the United States](https://mlsanity.com/paper/2210.08103/) (2022)
- [Back2Future: Leveraging Backfill Dynamics for Improving Real-time Predictions in Future](https://mlsanity.com/paper/2106.04420/) (2021)
- [Ensembles of Realistic Power Distribution Networks](https://mlsanity.com/paper/2208.06384/) (2022)
- [Continuously Updating Digital Twins using Large Language Models](https://mlsanity.com/paper/2506.12091/) (2025)
- [CALYPSO: Forecasting and Analyzing MRSA Infection Patterns with Community and Healthcare Transmission Dynamics](https://mlsanity.com/paper/2508.13548/) (2025)
- [Agentic Framework for Epidemiological Modeling](https://mlsanity.com/paper/2602.00299/) (2026)

## Built on by (most cited first, 1 of 1)

- [NeuronDiscover: Agent-in-Twin for Mechanistic Discovery in Neuronal Microenvironments with World Action Models](https://mlsanity.com/paper/2609.35338/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have shown promise in constructing mechanistic models from data. However, existing evaluations largely focus on simplified settings and fail to capture the complexity of real-world scientific modeling. In practice, such modeling often involves neural-integrated formulations, where a mechanistic model component and a neural network component are jointly constructed, leading to a significantly more complex search space. Motivated by this gap, we introduce the Neural-Integrated Mechanistic Modeling (NIMM) benchmark, which evaluates LLM-generated neural-integrated mechanistic models across three scientific domains. Experiments on NIMM reveal that existing LLM-based approaches struggle to effectively explore this complex space, resulting in limited search stability and solution quality. To address this challenge, we propose NIMMGen, a tree-guided agentic framework that enables diversified exploration via branch-level search and improves solutions through atomic model refinement. Extensive experiments demonstrate that NIMMGen achieves state-of-the-art performance on NIMM, significantly improving search stability and solution quality.