# Revealing economic facts: LLMs know more than they say

arXiv: https://arxiv.org/abs/2505.08662 · HTML: https://mlsanity.com/paper/2505.08662/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 37 references that are in the map.
- First posted on arXiv 13 May 2025.

## Builds on (most cited first, 21 of 21)

- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Probing Classifiers: Promises, Shortcomings, and Advances](https://mlsanity.com/paper/2102.12452/) (2021)
- [The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets](https://mlsanity.com/paper/2310.06824/) (2023)
- [MTEB: Massive Text Embedding Benchmark](https://mlsanity.com/paper/2210.07316/) (2022)
- [Learning from Noisy Labels with Deep Neural Networks: A Survey](https://mlsanity.com/paper/2007.08199/) (2020)
- [NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models](https://mlsanity.com/paper/2405.17428/) (2024)
- [Early-Learning Regularization Prevents Memorization of Noisy Labels](https://mlsanity.com/paper/2007.00151/) (2020)
- [Language Models Represent Space and Time](https://mlsanity.com/paper/2310.02207/) (2023)
- [LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations](https://mlsanity.com/paper/2410.02707/) (2024)
- [Deep Learning is Robust to Massive Label Noise](https://mlsanity.com/paper/1705.10694/) (2017)
- [Truth is Universal: Robust Detection of Lies in LLMs](https://mlsanity.com/paper/2407.12831/) (2024)
- [Uncertainty Estimation and Quantification for LLMs: A Simple Supervised Approach](https://mlsanity.com/paper/2404.15993/) (2024)
- [Micro-Estimates of Wealth for all Low- and Middle-Income Countries](https://mlsanity.com/paper/2104.07761/) (2021)
- [Prompt-Augmented Linear Probing: Scaling beyond the Limit of Few-shot In-Context Learners](https://mlsanity.com/paper/2212.10873/) (2022)
- [Robustness to Label Noise Depends on the Shape of the Noise Distribution in Feature Space](https://mlsanity.com/paper/2206.01106/) (2022)
- [On the Scaling Laws of Geographical Representation in Language Models](https://mlsanity.com/paper/2402.19406/) (2024)
- [Matrix Completion with Quantified Uncertainty through Low Rank Gaussian Copula](https://mlsanity.com/paper/2006.10829/) (2020)
- [More than Correlation: Do Large Language Models Learn Causal Representations of Space?](https://mlsanity.com/paper/2312.16257/) (2023)
- [What can LLM tell us about cities?](https://mlsanity.com/paper/2411.16791/) (2024)
- [Logistic Regression makes small LLMs strong and explainable "tens-of-shot" classifiers](https://mlsanity.com/paper/2408.03414/) (2024)

## Built on by (most cited first, 3 of 3)

- [Personalized Constitutionally-Aligned Agentic Superego: Secure AI Behavior Aligned to Diverse Human Values](https://mlsanity.com/paper/2506.13774/) (2025)
- [Inflation Attitudes of Large Language Models](https://mlsanity.com/paper/2512.14306/) (2025)
- [How AI Forecasts AI Jobs: Benchmarking LLM Predictions of Labor Market Changes](https://mlsanity.com/paper/2510.23358/) (2025)

## Abstract (arXiv)

We investigate whether the hidden states of large language models (LLMs) can be used to estimate and impute economic and financial statistics. Focusing on county-level (e.g. unemployment) and firm-level (e.g. total assets) variables, we show that a simple linear model trained on the hidden states of open-source LLMs outperforms the models' text outputs. This suggests that hidden states capture richer economic information than the responses of the LLMs reveal directly. A learning curve analysis indicates that only a few dozen labelled examples are sufficient for training. We also propose a transfer learning method that improves estimation accuracy without requiring any labelled data for the target variable. Finally, we demonstrate the practical utility of hidden-state representations in super-resolution and data imputation tasks.