# Am I Blue or Is My Hobby Counting Teardrops? Expression Leakage in Large Language Models as a Symptom of Irrelevancy Disruption

arXiv: https://arxiv.org/abs/2508.01708 · HTML: https://mlsanity.com/paper/2508.01708/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 21 references that are in the map.
- First posted on arXiv 3 Aug 2025.

## Builds on (most cited first, 12 of 12)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale](https://mlsanity.com/paper/2406.17557/) (2024)
- [Large Language Models Can Be Easily Distracted by Irrelevant Context](https://mlsanity.com/paper/2302.00093/) (2023)
- [TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classification](https://mlsanity.com/paper/2010.12421/) (2020)
- [Does Liking Yellow Imply Driving a School Bus? Semantic Leakage in Language Models](https://mlsanity.com/paper/2408.06518/) (2024)
- [Multilingual large language models leak human stereotypes across language boundaries](https://mlsanity.com/paper/2312.07141/) (2023)
- [Prompt Sentiment: The Catalyst for LLM Change](https://mlsanity.com/paper/2503.13510/) (2025)
- [Scaling Down Semantic Leakage: Investigating Associative Bias in Smaller Language Models](https://mlsanity.com/paper/2501.06638/) (2025)

## Built on by (most cited first, 1 of 1)

- [Are Language Models Sensitive to Morally Irrelevant Distractors?](https://mlsanity.com/paper/2602.09416/) (2026)

## Abstract (arXiv)

Large language models (LLMs) have advanced natural language processing (NLP) skills such as through next-token prediction and self-attention, but their ability to integrate broad context also makes them prone to incorporating irrelevant information. Prior work has focused on semantic leakage, bias introduced by semantically irrelevant context. In this paper, we introduce expression leakage, a novel phenomenon where LLMs systematically generate sentimentally charged expressions that are semantically unrelated to the input context. To analyse the expression leakage, we collect a benchmark dataset along with a scheme to automatically generate a dataset from free-form text from common-crawl. In addition, we propose an automatic evaluation pipeline that correlates well with human judgment, which accelerates the benchmarking by decoupling from the need of annotation for each analysed model. Our experiments show that, as the model scales in the parameter space, the expression leakage reduces within the same LLM family. On the other hand, we demonstrate that expression leakage mitigation requires specific care during the model building process, and cannot be mitigated by prompting. In addition, our experiments indicate that, when negative sentiment is injected in the prompt, it disrupts the generation process more than the positive sentiment, causing a higher expression leakage rate.