“All that Glitters”: Techniques for Evaluations with Unreliable Model and Human Annotations (2025.findings-naacl)
Copied to clipboard
| Challenge: | Using standard metrics in the presence of poor labels masks label and model quality . evaluation techniques accounting for unreliable labels reveal important flaws, including spurious correlations and nonrandom racial biases . |
| Approach: | They analyze human labels, GPT model ratings, and transformer encoder model ratings . they show that standard metrics in the presence of poor labels mask label and model quality . |
| Outcome: | The proposed methods mask label and model quality even in the presence of poor models. |
Similar Papers
Are LLMs Better than Reported? Detecting Label Errors and Mitigating Their Effect on Model Performance (2025.emnlp-main)
Copied to clipboard
| Challenge: | Recent advances in large language models (LLMs) offer new opportunities to enhance the annotation process, particularly for detecting label errors in existing datasets. |
| Approach: | They propose to use an ensemble of large language models to flag mislabeled examples by using an LLM-as-a-judge approach to detect label errors in existing datasets. |
| Outcome: | The proposed method improves label accuracy and consistency in large language models. |
GPT is Not an Annotator: The Necessity of Human Annotation in Fairness Benchmark Construction (2024.acl-long)
Copied to clipboard
| Challenge: | Current benchmarks for social biases have limitations in scope, grounding, quality and human effort required. |
| Approach: | They propose to use a language model to help with the development of bias benchmarks . they extend previous work to a new community and set of biases: the Jewish community and antisemitism . |
| Outcome: | The proposed LLM does not perform well on the Jewish community and antisemitism task. |
Revisiting the Gold Standard: Grounding Summarization Evaluation with Robust Human Evaluation (2023.acl-long)
Copied to clipboard
Yixin Liu, Alex Fabbri, Pengfei Liu, Yilun Zhao, Linyong Nan, Ruilin Han, Simeng Han, Shafiq Joty, Chien-Sheng Wu, Caiming Xiong, Dragomir Radev
| Challenge: | Existing studies for summarization evaluation exhibit low inter-annotator agreement or lack scale. |
| Approach: | They propose a modified summarization salience protocol based on fine-grained semantic units and a robust summarizing evaluation benchmark. |
| Outcome: | The proposed protocol is based on fine-grained semantic units and allows for high inter-annotator agreement. |
Evaluating Large Language Models on Wikipedia-Style Survey Generation (2024.findings-acl)
Copied to clipboard
Fan Gao, Hang Jiang, Rui Yang, Qingcheng Zeng, Jinghui Lu, Moritz Blum, Tianwei She, Yuang Jiang, Irene Li
| Challenge: | Recent studies have shown that large language models can perform well in general tasks, but their effectiveness and limitations in domainspecific tasks remain unclear. |
| Approach: | They examine the proficiency of Large Language Models (LLMs) in generating succinct survey articles specific to the niche field of NLP in computer science. |
| Outcome: | The LLMs perform better in generating succinct survey articles specific to the niche field of NLP in computer science, compared to human-authored surveys, but they exhibit bias in evaluation. |
All That’s ‘Human’ Is Not Gold: Evaluating Human Evaluation of Generated Text (2021.acl-long)
Copied to clipboard
| Challenge: | evaluators distinguish between human- and machine-authored text in three domains without training . evals' accuracy improved up to 55%, but it did not significantly improve across the three domain. |
| Approach: | They examine the role untrained human evaluations play in NLG evaluation and propose ways to improve their evaluations. |
| Outcome: | The evaluators distinguished between human- and machine-authored text at random chance level without training, but their accuracy did not improve across the three domains. |
The “Problem” of Human Label Variation: On Ground Truth in Data, Modeling and Evaluation (2022.emnlp-main)
Copied to clipboard
| Challenge: | a paper argues that human label variation impacts all stages of the ML pipeline . human label variations are often considered noise due to disagreement, subjectivity in annotation or multiple plausible answers. |
| Approach: | They propose to reconcile different notions of human label variation and propose a repository of publicly-available datasets with un-aggregated labels. |
| Outcome: | The proposed approaches are compared with publicly available datasets with un-aggregated labels and identify gaps. |
VariErr NLI: Separating Annotation Error from Human Label Variation (2024.acl-long)
Copied to clipboard
| Challenge: | Existing work on label variation and annotation errors has focused on them in isolation. |
| Approach: | They propose a 2-round annotation procedure to separate human label variation from annotation errors by pairing valid explanations with annotators' validations. |
| Outcome: | The proposed procedure is based on the NLI task in English and contains 7,732 valid judgements on 1,933 explanations for 500 re-annotated items. |
How Many Ratings per Item are Necessary for Reliable Significance Testing? (2026.findings-eacl)
Copied to clipboard
| Challenge: | Existing methods for estimating model reliability are based on a few output responses per item. |
| Approach: | They propose a method to determine whether an existing dataset has enough responses per item to assure reliable null hypothesis statistical testing. |
| Outcome: | The proposed method can help researchers make better decisions about how to collect data for AI evaluation. |
Humans Hallucinate Too: Language Models Identify and Correct Subjective Annotation Errors With Label-in-a-Haystack Prompts (2025.emnlp-main)
Copied to clipboard
Georgios Chochlakis, Peter Wu, Tikka Arjun Singh Bedi, Marcus Ma, Kristina Lerman, Shrikanth Narayanan
| Challenge: | Existing approaches to model complex subjective tasks in natural language are limited by significant variation in annotations. |
| Approach: | They propose a simple in-context learning binary filtering baseline that estimates the reasonableness of a document-label pair. |
| Outcome: | The proposed approach can be integrated into annotation pipelines to enhance signal-to-noise ratios. |
Refined Assessment for Translation Evaluation: Rethinking Machine Translation Evaluation in the Era of Human-Level Systems (2025.findings-emnlp)
Copied to clipboard
Dmitry Popov, Vladislav Negodin, Ekaterina Enikeeva, Iana Matrosova, Nikolay Karpachev, Max Ryabinin
| Challenge: | Currently, traditional evaluation methods struggle to detect subtle translation errors. |
| Approach: | They propose to use a dataset of human evaluations for English–Russian translations created by professional linguists to enable consistent and rich annotation. |
| Outcome: | The proposed protocol allows expert assessments without time pressure to yield substantially different results from standard evaluations. |