English
Related papers

Related papers: The Authenticity Gap in Human Evaluation

200 papers

Modern Large Language Models (LLMs) have showcased remarkable prowess in various tasks necessitating sophisticated cognitive behaviors. Nevertheless, a paradoxical performance discrepancy is observed, where these models underperform in…

Computation and Language · Computer Science 2024-04-05 Yuchen Fan , Yantao Liu , Zijun Yao , Jifan Yu , Lei Hou , Juanzi Li

Modern affective computing systems rely heavily on datasets with human-annotated emotion labels, for training and evaluation. However, human annotations are expensive to obtain, sensitive to study design, and difficult to quality control,…

Computation and Language · Computer Science 2024-12-12 Minxue Niu , Yara El-Tawil , Amrit Romana , Emily Mower Provost

The quality of meeting summaries generated by natural language generation (NLG) systems is hard to measure automatically. Established metrics such as ROUGE and BERTScore have a relatively low correlation with human judgments and fail to…

Computation and Language · Computer Science 2025-02-19 Frederic Kirstein , Terry Ruas , Bela Gipp

Precisely assessing the progress in natural language generation (NLG) tasks is challenging, and human evaluation to establish a preference in a model's output over another is often necessary. However, human evaluation is usually costly,…

Computation and Language · Computer Science 2022-11-10 Philippe Laban , Chien-Sheng Wu , Wenhao Liu , Caiming Xiong

In this study, we analyze automatic evaluation metrics for Natural Language Generation (NLG), specifically task-agnostic metrics and human-aligned metrics. Task-agnostic metrics, such as Perplexity, BLEU, BERTScore, are cost-effective and…

Computation and Language · Computer Science 2023-05-29 Iftitahu Ni'mah , Meng Fang , Vlado Menkovski , Mykola Pechenizkiy

Summarization systems are ultimately evaluated by human annotators and raters. Usually, annotators and raters do not reflect the demographics of end users, but are recruited through student populations or crowdsourcing platforms with skewed…

Computation and Language · Computer Science 2021-10-12 Anna Jørgensen , Anders Søgaard

Progress on many Natural Language Processing (NLP) tasks, such as text classification, is driven by objective, reproducible and scalable evaluation via publicly available benchmarks. However, these are not always representative of…

Computation and Language · Computer Science 2022-11-11 Viktor Schlegel , Erick Mendez-Guzman , Riza Batista-Navarro

The use of large language models like ChatGPT in code review offers promising efficiency gains but also raises concerns about correctness and safety. Existing evaluation methods for code review generation either rely on automatic…

Software Engineering · Computer Science 2025-12-18 Robert Heumüller , Frank Ortmeier

Conversational systems are now capable of producing impressive and generally relevant responses. However, we have no visibility nor control of the socio-emotional strategies behind state-of-the-art Large Language Models (LLMs), which poses…

Computation and Language · Computer Science 2024-12-09 Lorraine Vanel , Ariel R. Ramos Vela , Alya Yacoubi , Chloé Clavel

Serendipity plays a pivotal role in enhancing user satisfaction within recommender systems, yet its evaluation poses significant challenges due to its inherently subjective nature and conceptual ambiguity. Current algorithmic approaches…

Information Retrieval · Computer Science 2025-07-24 Li Kang , Yuhan Zhao , Li Chen

As generative AI models such as large language models (LLMs) become more pervasive, ensuring the safety, robustness, and overall trustworthiness of these systems is paramount. However, AI is currently facing a reproducibility crisis driven…

Machine Learning · Computer Science 2026-05-14 Deepak Pandita , Flip Korn , Chris Welty , Christopher M. Homan

Traditional LLM alignment methods are vulnerable to heterogeneity in human preferences. Fitting a na\"ive probabilistic model to pairwise comparison data (say over prompt-completion pairs) yields an inconsistent estimate of the…

Artificial Intelligence · Computer Science 2025-10-30 Ali Aouad , Aymane El Gadarri , Vivek F. Farias

Validating evaluation metrics for NLG typically relies on expensive and time-consuming human annotations, which predominantly exist only for English datasets. We propose \textit{LLM as a Meta-Judge}, a scalable framework that utilizes LLMs…

Computation and Language · Computer Science 2026-03-11 Lukáš Eigler , Jindřich Libovický , David Hurych

Language serves as a powerful tool for the manifestation of societal belief systems. In doing so, it also perpetuates the prevalent biases in our society. Gender bias is one of the most pervasive biases in our society and is seen in online…

Computation and Language · Computer Science 2023-10-27 Rishav Hada , Agrima Seth , Harshita Diddee , Kalika Bali

The well-studied problem of statistical rank aggregation has been applied to comparing sports teams, information retrieval, and most recently to data generated by human judgment. Such human-generated rankings may be substantially different…

Information Retrieval · Computer Science 2014-11-05 Andrew Mao , Hossein Azari Soufiani , Yiling Chen , David C. Parkes

Natural Language Generation (NLG) evaluation is a multifaceted task requiring assessment of multiple desirable criteria, e.g., fluency, coherency, coverage, relevance, adequacy, overall quality, etc. Across existing datasets for 6 NLG…

Computation and Language · Computer Science 2021-09-14 Ananya B. Sai , Tanay Dixit , Dev Yashpal Sheth , Sreyas Mohan , Mitesh M. Khapra

A common practice in building NLP datasets, especially using crowd-sourced annotations, involves obtaining multiple annotator judgements on the same data instances, which are then flattened to produce a single "ground truth" label or score,…

Computation and Language · Computer Science 2021-10-13 Vinodkumar Prabhakaran , Aida Mostafazadeh Davani , Mark Díaz

In the context of text classification, the financial burden of annotation exercises for creating training data is a critical issue. Active learning techniques, particularly those rooted in uncertainty sampling, offer a cost-effective…

Computation and Language · Computer Science 2024-06-19 Hamidreza Rouzegar , Masoud Makrehchi

Human evaluation is viewed as a reliable evaluation method for NLG which is expensive and time-consuming. To save labor and costs, researchers usually perform human evaluation on a small subset of data sampled from the whole dataset in…

Computation and Language · Computer Science 2024-06-13 Jie Ruan , Xiao Pu , Mingqi Gao , Xiaojun Wan , Yuesheng Zhu

Evaluating the quality of search, ranking and RAG systems traditionally requires a significant number of human relevance annotations. In recent times, several deployed systems have explored the usage of Large Language Models (LLMs) as…

Machine Learning · Computer Science 2026-01-27 Abhishek Divekar , Anirban Majumder