Humans Hallucinate Too: Language Models Identify and Correct Subjective Annotation Errors With Label-in-a-Haystack Prompts
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908642586918912 |
|---|---|
| author | Chochlakis, Georgios Wu, Peter Bedi, Arjun Ma, Marcus Lerman, Kristina Narayanan, Shrikanth |
| author_facet | Chochlakis, Georgios Wu, Peter Bedi, Arjun Ma, Marcus Lerman, Kristina Narayanan, Shrikanth |
| contents | Modeling complex subjective tasks in Natural Language Processing, such as recognizing emotion and morality, is considerably challenging due to significant variation in human annotations. This variation often reflects reasonable differences in semantic interpretations rather than mere noise, necessitating methods to distinguish between legitimate subjectivity and error. We address this challenge by exploring label verification in these contexts using Large Language Models (LLMs). First, we propose a simple In-Context Learning binary filtering baseline that estimates the reasonableness of a document-label pair. We then introduce the Label-in-a-Haystack setting: the query and its label(s) are included in the demonstrations shown to LLMs, which are prompted to predict the label(s) again, while receiving task-specific instructions (e.g., emotion recognition) rather than label copying. We show how the failure to copy the label(s) to the output of the LLM are task-relevant and informative. Building on this, we propose the Label-in-a-Haystack Rectification (LiaHR) framework for subjective label correction: when the model outputs diverge from the reference gold labels, we assign the generated labels to the example instead of discarding it. This approach can be integrated into annotation pipelines to enhance signal-to-noise ratios. Comprehensive analyses, human evaluations, and ecological validity studies verify the utility of LiaHR for label correction. Code is available at https://github.com/gchochla/liahr. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_17222 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Humans Hallucinate Too: Language Models Identify and Correct Subjective Annotation Errors With Label-in-a-Haystack Prompts Chochlakis, Georgios Wu, Peter Bedi, Arjun Ma, Marcus Lerman, Kristina Narayanan, Shrikanth Computation and Language Modeling complex subjective tasks in Natural Language Processing, such as recognizing emotion and morality, is considerably challenging due to significant variation in human annotations. This variation often reflects reasonable differences in semantic interpretations rather than mere noise, necessitating methods to distinguish between legitimate subjectivity and error. We address this challenge by exploring label verification in these contexts using Large Language Models (LLMs). First, we propose a simple In-Context Learning binary filtering baseline that estimates the reasonableness of a document-label pair. We then introduce the Label-in-a-Haystack setting: the query and its label(s) are included in the demonstrations shown to LLMs, which are prompted to predict the label(s) again, while receiving task-specific instructions (e.g., emotion recognition) rather than label copying. We show how the failure to copy the label(s) to the output of the LLM are task-relevant and informative. Building on this, we propose the Label-in-a-Haystack Rectification (LiaHR) framework for subjective label correction: when the model outputs diverge from the reference gold labels, we assign the generated labels to the example instead of discarding it. This approach can be integrated into annotation pipelines to enhance signal-to-noise ratios. Comprehensive analyses, human evaluations, and ecological validity studies verify the utility of LiaHR for label correction. Code is available at https://github.com/gchochla/liahr. |
| title | Humans Hallucinate Too: Language Models Identify and Correct Subjective Annotation Errors With Label-in-a-Haystack Prompts |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2505.17222 |