# Knowing the Facts but Choosing the Shortcut: Understanding How Large Language Models Compare Entities

arXiv: https://arxiv.org/abs/2510.16815 · HTML: https://mlsanity.com/paper/2510.16815/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 31 references that are in the map.
- First posted on arXiv 19 Oct 2025.

## Builds on (most cited first, 19 of 19)

- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Measuring and Narrowing the Compositionality Gap in Language Models](https://mlsanity.com/paper/2210.03350/) (2022)
- [ConceptNet 5.5: An Open Multilingual Graph of General Knowledge](https://mlsanity.com/paper/1612.03975/) (2016)
- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Reasoning Models Don't Always Say What They Think](https://mlsanity.com/paper/2505.05410/) (2025)
- [Faithful Chain-of-Thought Reasoning](https://mlsanity.com/paper/2301.13379/) (2023)
- [Physics of Language Models: Part 3.1, Knowledge Storage and Extraction](https://mlsanity.com/paper/2309.14316/) (2023)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [Physics of Language Models: Part 3.2, Knowledge Manipulation](https://mlsanity.com/paper/2309.14402/) (2023)
- [Inducing Relational Knowledge from BERT](https://mlsanity.com/paper/1911.12753/) (2019)
- [Number Cookbook: Number Understanding of Language Models and How to Improve It](https://mlsanity.com/paper/2411.03766/) (2024)
- [Investigating Non-Transitivity in LLM-as-a-Judge](https://mlsanity.com/paper/2502.14074/) (2025)
- [NumeroLogic: Number Encoding for Enhanced LLMs' Numerical Reasoning](https://mlsanity.com/paper/2404.00459/) (2024)
- [Emergent Representations of Program Semantics in Language Models Trained on Programs](https://mlsanity.com/paper/2305.11169/) (2023)
- [Calibrating Reasoning in Language Models with Internal Consistency](https://mlsanity.com/paper/2405.18711/) (2024)
- [Machine Reading, Fast and Slow: When Do Models "Understand" Language?](https://mlsanity.com/paper/2209.07430/) (2022)
- [Stochastic LLMs do not Understand Language: Towards Symbolic, Explainable and Ontologically Based LLMs](https://mlsanity.com/paper/2309.05918/) (2023)
- [Ranking Entities along Conceptual Space Dimensions with LLMs: An Analysis of Fine-Tuning Strategies](https://mlsanity.com/paper/2402.15337/) (2024)

## Built on by (most cited first, 2 of 2)

- [Detecting Hallucinations in Retrieval-Augmented Generation via Semantic-level Internal Reasoning Graph](https://mlsanity.com/paper/2601.03052/) (2026)
- [Pretraining Exposure Explains Popularity Judgments in Large Language Models](https://mlsanity.com/paper/2605.12382/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) are increasingly used for knowledge-based reasoning tasks, yet understanding when they rely on genuine knowledge versus superficial heuristics remains challenging. We investigate this question through entity comparison tasks by asking models to compare entities along numerical attributes (e.g., ``Which river is longer, the Danube or the Nile?''), which offer clear ground truth for systematic analysis. Despite having sufficient numerical knowledge to answer correctly, LLMs frequently make predictions that contradict this knowledge. We identify three heuristic biases that strongly influence model predictions: entity popularity, mention order, and semantic co-occurrence. For smaller models, a simple logistic regression using only these surface cues predicts model choices more accurately than the model's own numerical predictions, suggesting heuristics largely override principled reasoning. Crucially, we find that larger models (32B parameters) selectively rely on numerical knowledge when it is more reliable, while smaller models (7--8B parameters) show no such discrimination, which explains why larger models outperform smaller ones even when the smaller models possess more accurate knowledge. Chain-of-thought prompting steers all models towards using the numerical features across all model sizes.