Enregistré dans:
| Auteurs principaux: | , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2505.08253 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866918018672492544 |
|---|---|
| author | Miller, Justin K Tang, Wenjia |
| author_facet | Miller, Justin K Tang, Wenjia |
| contents | As generative AI becomes increasingly embedded in everyday workflows, it is important to evaluate its performance in ways that reflect real-world usage rather than abstract notions of intelligence. Unlike many existing benchmarks that assess general intelligence, our approach focuses on real-world utility, evaluating how well models support users in everyday tasks. While current benchmarks emphasize code generation or factual recall, users rely on AI for a much broader range of activities-from writing assistance and summarization to citation formatting and stylistic feedback. In this paper, we analyze large-scale survey data and usage logs to identify six core capabilities that represent how people commonly use Large Language Models (LLMs): Summarization, Technical Assistance, Reviewing Work, Data Structuring, Generation, and Information Retrieval. We then assess the extent to which existing benchmarks cover these capabilities, revealing significant gaps in coverage, efficiency measurement, and interpretability. Drawing on this analysis, we use human-centered criteria to identify gaps in how well current benchmarks reflect common usage that is grounded in five practical criteria: coherence, accuracy, clarity, relevance, and efficiency. For four of the six capabilities, we identify the benchmarks that best align with real-world tasks and use them to compare leading models. We find that Google Gemini outperforms other models-including OpenAI's GPT, xAI's Grok, Meta's LLaMA, Anthropic's Claude, DeepSeek, and Qwen from Alibaba-on these utility-focused metrics. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_08253 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Evaluating LLM Metrics Through Real-World Capabilities Miller, Justin K Tang, Wenjia Artificial Intelligence I.2.7 As generative AI becomes increasingly embedded in everyday workflows, it is important to evaluate its performance in ways that reflect real-world usage rather than abstract notions of intelligence. Unlike many existing benchmarks that assess general intelligence, our approach focuses on real-world utility, evaluating how well models support users in everyday tasks. While current benchmarks emphasize code generation or factual recall, users rely on AI for a much broader range of activities-from writing assistance and summarization to citation formatting and stylistic feedback. In this paper, we analyze large-scale survey data and usage logs to identify six core capabilities that represent how people commonly use Large Language Models (LLMs): Summarization, Technical Assistance, Reviewing Work, Data Structuring, Generation, and Information Retrieval. We then assess the extent to which existing benchmarks cover these capabilities, revealing significant gaps in coverage, efficiency measurement, and interpretability. Drawing on this analysis, we use human-centered criteria to identify gaps in how well current benchmarks reflect common usage that is grounded in five practical criteria: coherence, accuracy, clarity, relevance, and efficiency. For four of the six capabilities, we identify the benchmarks that best align with real-world tasks and use them to compare leading models. We find that Google Gemini outperforms other models-including OpenAI's GPT, xAI's Grok, Meta's LLaMA, Anthropic's Claude, DeepSeek, and Qwen from Alibaba-on these utility-focused metrics. |
| title | Evaluating LLM Metrics Through Real-World Capabilities |
| topic | Artificial Intelligence I.2.7 |
| url | https://arxiv.org/abs/2505.08253 |