# Auxiliary task demands mask the capabilities of smaller language models

arXiv: https://arxiv.org/abs/2404.02418 · HTML: https://mlsanity.com/paper/2404.02418/

## Key facts

- Cited by 54 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 50 references that are in the map.
- First posted on arXiv 3 Apr 2024.

## Builds on (most cited first, 30 of 31)

- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Large Language Models are Zero-Shot Reasoners](https://mlsanity.com/paper/2205.11916/) (2022)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [Pythia: A Suite for Analyzing Large Language Models Across Training and Scaling](https://mlsanity.com/paper/2304.01373/) (2023)
- [The LAMBADA dataset: Word prediction requiring a broad discourse context](https://mlsanity.com/paper/1606.06031/) (2016)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [Grokking: Generalization Beyond Overfitting on Small Algorithmic Datasets](https://mlsanity.com/paper/2201.02177/) (2022)
- [BLiMP: The Benchmark of Linguistic Minimal Pairs for English](https://mlsanity.com/paper/1912.00582/) (2019)
- [Are Emergent Abilities of Large Language Models a Mirage?](https://mlsanity.com/paper/2304.15004/) (2023)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)
- [Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies](https://mlsanity.com/paper/1611.01368/) (2016)
- [AI and the Everything in the Whole Wide World Benchmark](https://mlsanity.com/paper/2111.15366/) (2021)
- [Targeted Syntactic Evaluation of Language Models](https://mlsanity.com/paper/1808.09031/) (2018)
- [Emergent Analogical Reasoning in Large Language Models](https://mlsanity.com/paper/2212.09196/) (2022)
- [What Will it Take to Fix Benchmarking in Natural Language Understanding?](https://mlsanity.com/paper/2104.02145/) (2021)
- [Language models show human-like content effects on reasoning tasks](https://mlsanity.com/paper/2207.07051/) (2022)
- [Political Compass or Spinning Arrow? Towards More Meaningful Evaluations for Values and Opinions in Large Language Models](https://mlsanity.com/paper/2402.16786/) (2024)
- [Thinking Fast and Slow in Large Language Models](https://mlsanity.com/paper/2212.05206/) (2022)
- [How Can We Accelerate Progress Towards Human-like Linguistic Generalization?](https://mlsanity.com/paper/2005.00955/) (2020)
- [The Generative AI Paradox: "What It Can Create, It May Not Understand"](https://mlsanity.com/paper/2311.00059/) (2023)
- [Language Model Behavior: A Comprehensive Survey](https://mlsanity.com/paper/2303.11504/) (2023)
- [Evaluating Cognitive Maps and Planning in Large Language Models with CogEval](https://mlsanity.com/paper/2309.15129/) (2023)
- [minicons: Enabling Flexible Behavioral and Representational Analyses of Transformer Language Models](https://mlsanity.com/paper/2203.13112/) (2022)
- [Language models align with human judgments on key grammatical constructions](https://mlsanity.com/paper/2402.01676/) (2024)
- [Prompting is not a substitute for probability measurements in large language models](https://mlsanity.com/paper/2305.13264/) (2023)
- [Can language models handle recursively nested grammatical structures? A case study on comparing models and humans](https://mlsanity.com/paper/2210.15303/) (2022)
- [The Dangers of Underclaiming: Reasons for Caution When Reporting How NLP Systems Fail](https://mlsanity.com/paper/2110.08300/) (2021)
- [On the Stepwise Nature of Self-Supervised Learning](https://mlsanity.com/paper/2303.15438/) (2023)
- [Beyond Probabilities: Unveiling the Misalignment in Evaluating Large Language Models](https://mlsanity.com/paper/2402.13887/) (2024)
- [Predictions from language models for multiple-choice tasks are not robust under variation of scoring methods](https://mlsanity.com/paper/2403.00998/) (2024)

## Built on by (most cited first, 30 of 54)

- [Are Large Language Models Consistent over Value-laden Questions?](https://mlsanity.com/paper/2407.02996/) (2024)
- [One fish, two fish, but not the whole sea: Alignment reduces language models' conceptual diversity](https://mlsanity.com/paper/2411.04427/) (2024)
- [Elements of World Knowledge (EWoK): A Cognition-Inspired Framework for Evaluating Basic World Knowledge in Language Models](https://mlsanity.com/paper/2405.09605/) (2024)
- [Social Bias Evaluation for Large Language Models Requires Prompt Variations](https://mlsanity.com/paper/2407.03129/) (2024)
- [Language Models Fail to Introspect About Their Knowledge of Language](https://mlsanity.com/paper/2503.07513/) (2025)
- [LINGOLY: A Benchmark of Olympiad-Level Linguistic Reasoning Puzzles in Low-Resource and Extinct Languages](https://mlsanity.com/paper/2406.06196/) (2024)
- [On Benchmarking Human-Like Intelligence in Machines](https://mlsanity.com/paper/2502.20502/) (2025)
- [Language Models as Models of Language](https://mlsanity.com/paper/2408.07144/) (2024)
- [Re-evaluating Theory of Mind evaluation in large language models](https://mlsanity.com/paper/2502.21098/) (2025)
- [Robust Pronoun Fidelity with English LLMs: Are they Reasoning, Repeating, or Just Biased?](https://mlsanity.com/paper/2404.03134/) (2024)
- [LLMs as Models for Analogical Reasoning](https://mlsanity.com/paper/2406.13803/) (2024)
- [What Can String Probability Tell Us About Grammaticality?](https://mlsanity.com/paper/2510.16227/) (2025)
- [Aligned but Blind: Alignment Increases Implicit Bias by Reducing Awareness of Race](https://mlsanity.com/paper/2506.00253/) (2025)
- [Evidence from counterfactual tasks supports emergent analogical reasoning in large language models](https://mlsanity.com/paper/2404.13070/) (2024)
- [A Survey on Collaborative Mechanisms Between Large and Small Language Models](https://mlsanity.com/paper/2505.07460/) (2025)
- [DevBench: A multimodal developmental benchmark for language learning](https://mlsanity.com/paper/2406.10215/) (2024)
- [Language Model Behavioral Phases are Consistent Across Architecture, Training Data, and Scale](https://mlsanity.com/paper/2510.24963/) (2025)
- [Searching for Structure: Investigating Emergent Communication with Large Language Models](https://mlsanity.com/paper/2412.07646/) (2024)
- [Levels of Analysis for Large Language Models](https://mlsanity.com/paper/2503.13401/) (2025)
- [When Do Large Language Models Exhibit Unsolicited Deception?](https://mlsanity.com/paper/2504.00285/) (2025)
- [WinoPron: Revisiting English Winogender Schemas for Consistency, Coverage, and Grammatical Case](https://mlsanity.com/paper/2409.05653/) (2024)
- [Leaving the barn door open for Clever Hans: Simple features predict LLM benchmark answers](https://mlsanity.com/paper/2410.11672/) (2024)
- [Dissecting the Ullman Variations with a SCALPEL: Why do LLMs fail at Trivial Alterations to the False Belief Task?](https://mlsanity.com/paper/2406.14737/) (2024)
- [Random Scaling of Emergent Capabilities](https://mlsanity.com/paper/2502.17356/) (2025)
- [Experimental Pragmatics with Machines: Testing LLM Predictions for the Inferences of Plain and Embedded Disjunctions](https://mlsanity.com/paper/2405.05776/) (2024)
- [Bigram Subnetworks: Mapping to Next Tokens in Transformer Language Models](https://mlsanity.com/paper/2504.15471/) (2025)
- [Large Language Model probabilities cannot distinguish between possible and impossible language](https://mlsanity.com/paper/2509.15114/) (2025)
- [Anthropocentric bias in language model evaluation](https://mlsanity.com/paper/2407.03859/) (2024)
- [Incremental Sentence Processing Mechanisms in Autoregressive Transformer Language Models](https://mlsanity.com/paper/2412.05353/) (2024)
- [Understanding Common Ground Misalignment in Goal-Oriented Dialog: A Case-Study with Ubuntu Chat Logs](https://mlsanity.com/paper/2503.12370/) (2025)

## Abstract (arXiv)

Developmental psychologists have argued about when cognitive capacities such as language understanding or theory of mind emerge. These debates often hinge on the concept of "task demands" -- the auxiliary challenges associated with performing a particular evaluation -- that may mask the child's underlying ability. The same issues arise when measuring the capacities of language models (LMs): performance on a task is a function of the model's underlying knowledge, combined with the model's ability to interpret and perform the task given its available resources. Here, we show that for analogical reasoning, reflective reasoning, word prediction, and grammaticality judgments, evaluation methods with greater task demands yield lower performance than evaluations with reduced demands. This "demand gap" is most pronounced for models with fewer parameters and less training data. Our results illustrate that LM performance should not be interpreted as a direct indication of intelligence (or lack thereof), but as a reflection of capacities seen through the lens of researchers' design choices.