Entity · paper

Language Models Compare Quantities Using Number-specific and Unit-specific Heuristics

paperactivelanguage-models-compare-quantities-using-number-specific-and-unit-specific-heuristics-68b57b7a·1 events·first seen Jun 3, 2026

Aliases: Language Models Compare Quantities Using Number-specific and Unit-specific Heuristics

More like this (12)

Language Models as Measurement Apparatus for Culture Arithmetic Pedagogy for Language Models The Illusion of Equivalency: Statistical Characterization of Quantization Effects in LLMs Reasoning Language Models Language Modeling Loss Prompt Design at Scale: How Format, Instruction Count, and Context Length Shape Instruction Adherence and Hallucination in Large Language Models Scaling Laws for Neural Language Models Decomposing Factual Sycophancy in Language Models: How Size and Instruction Tuning Shape Robustness 1B-scale language models Rate-Utility Frontiers for Language Encodings: Comparing Tokens, Bytes, and Pixels Under Controlled Linguistic Content Uncertainty Quantification for Computer-Use Agents: A Benchmark across Vision-Language Models and GUI Grounding Datasets Instruction-Tuned Models Locally Reuse Human Syntax More Than Humans Do

Recent events (1)

4arXiv · cs.CL·Jun 3, 2026·source ↗

LMs compare quantities via number and unit heuristics rather than exact conversion

A new arXiv paper investigates how language models handle quantity comparisons involving measurement units (e.g., 110 cm vs. 1.2 m) across multiple unit systems. The authors find that accuracy degrades near comparison boundaries and that LM behavior is explained by linear surrogate models using numerical-difference and unit-scale-difference cues. Causal interventions confirm these heuristics are mechanistically operative, suggesting LMs do not perform exact unit conversion but instead rely on a bag of numerical and unit-scale shortcuts.

Evaluation and Benchmarking AI Safety Research Language Models Compare Quantities Using Number-specific and Unit-specific Heuristics