# From Form(s) to Meaning: Probing the Semantic Depths of Language Models Using Multisense Consistency

arXiv: https://arxiv.org/abs/2404.12145 · HTML: https://mlsanity.com/paper/2404.12145/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 71 of its 117 references that are in the map.
- First posted on arXiv 18 Apr 2024.

## Builds on (most cited first, 30 of 71)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding](https://mlsanity.com/paper/1804.07461/) (2018)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [Transformer Feed-Forward Layers Are Key-Value Memories](https://mlsanity.com/paper/2012.14913/) (2020)
- [Atlas: Few-shot Learning with Retrieval Augmented Language Models](https://mlsanity.com/paper/2208.03299/) (2022)
- [XNLI: Evaluating Cross-lingual Sentence Representations](https://mlsanity.com/paper/1809.05053/) (2018)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [XTREME: A Massively Multilingual Multi-task Benchmark for Evaluating Cross-lingual Generalization](https://mlsanity.com/paper/2003.11080/) (2020)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [The Reversal Curse: LLMs trained on "A is B" fail to learn "B is A"](https://mlsanity.com/paper/2309.12288/) (2023)
- [PAWS: Paraphrase Adversaries from Word Scrambling](https://mlsanity.com/paper/1904.01130/) (2019)
- [Measuring and Improving Consistency in Pretrained Language Models](https://mlsanity.com/paper/2102.01017/) (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)
- [XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning](https://mlsanity.com/paper/2005.00333/) (2020)
- [The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants](https://mlsanity.com/paper/2308.16884/) (2023)
- [State of What Art? A Call for Multi-Prompt LLM Evaluation](https://mlsanity.com/paper/2401.00595/) (2023)
- [PAWS-X: A Cross-lingual Adversarial Dataset for Paraphrase Identification](https://mlsanity.com/paper/1908.11828/) (2019)
- [Negated and Misprimed Probes for Pretrained Language Models: Birds Can Talk, But Cannot Fly](https://mlsanity.com/paper/1911.03343/) (2019)
- [Experience Grounds Language](https://mlsanity.com/paper/2004.10151/) (2020)
- [Adversarial GLUE: A Multi-Task Benchmark for Robustness Evaluation of Language Models](https://mlsanity.com/paper/2111.02840/) (2021)
- [XGLUE: A New Benchmark Dataset for Cross-lingual Pre-training, Understanding and Generation](https://mlsanity.com/paper/2004.01401/) (2020)
- [Probing Neural Network Comprehension of Natural Language Arguments](https://mlsanity.com/paper/1907.07355/) (2019)
- [Colorless green recurrent networks dream hierarchically](https://mlsanity.com/paper/1803.11138/) (2018)
- [Embers of Autoregression: Understanding Large Language Models Through the Problem They are Trained to Solve](https://mlsanity.com/paper/2309.13638/) (2023)
- [Are We Modeling the Task or the Annotator? An Investigation of Annotator Bias in Natural Language Understanding Datasets](https://mlsanity.com/paper/1908.07898/) (2019)
- [Sources of Hallucination by Large Language Models on Inference Tasks](https://mlsanity.com/paper/2305.14552/) (2023)
- [XTREME-R: Towards More Challenging and Nuanced Multilingual Evaluation](https://mlsanity.com/paper/2104.07412/) (2021)

## Built on by (most cited first, 7 of 7)

- [Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges](https://mlsanity.com/paper/2406.12624/) (2024)
- [MultiLoKo: a multilingual local knowledge benchmark for LLMs spanning 31 languages](https://mlsanity.com/paper/2504.10356/) (2025)
- [Towards Rationality in Language and Multimodal Agents: A Survey](https://mlsanity.com/paper/2406.00252/) (2024)
- [Consistency in Language Models: Current Landscape, Challenges, and Future Directions](https://mlsanity.com/paper/2505.00268/) (2025)
- [How LLMs Comprehend Temporal Meaning in Narratives: A Case Study in Cognitive Evaluation of LLMs](https://mlsanity.com/paper/2507.14307/) (2025)
- [Form and Meaning in Intrinsic Multilingual Evaluations](https://mlsanity.com/paper/2601.10580/) (2026)
- [MERGE: Minimal Expression-Replacement GEneralization Test for Natural Language Inference](https://mlsanity.com/paper/2510.24295/) (2025)

## Abstract (arXiv)

The staggering pace with which the capabilities of large language models (LLMs) are increasing, as measured by a range of commonly used natural language understanding (NLU) benchmarks, raises many questions regarding what "understanding" means for a language model and how it compares to human understanding. This is especially true since many LLMs are exclusively trained on text, casting doubt on whether their stellar benchmark performances are reflective of a true understanding of the problems represented by these benchmarks, or whether LLMs simply excel at uttering textual forms that correlate with what someone who understands the problem would say. In this philosophically inspired work, we aim to create some separation between form and meaning, with a series of tests that leverage the idea that world understanding should be consistent across presentational modes - inspired by Fregean senses - of the same meaning. Specifically, we focus on consistency across languages as well as paraphrases. Taking GPT-3.5 as our object of study, we evaluate multisense consistency across five different languages and various tasks. We start the evaluation in a controlled setting, asking the model for simple facts, and then proceed with an evaluation on four popular NLU benchmarks. We find that the model's multisense consistency is lacking and run several follow-up analyses to verify that this lack of consistency is due to a sense-dependent task understanding. We conclude that, in this aspect, the understanding of LLMs is still quite far from being consistent and human-like, and deliberate on how this impacts their utility in the context of learning about human language and understanding.