Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866911350063628288 |
|---|---|
| author | Kim, Doyoung Ren, Zhiwei Hao, Jie Sun, Zhongkai Wang, Lichao Ma, Xiyao Ye, Zack Han, Xu Yin, Jun Ji, Heng Shen, Wei Fan, Xing Yao, Benjamin Guo, Chenlei |
| author_facet | Kim, Doyoung Ren, Zhiwei Hao, Jie Sun, Zhongkai Wang, Lichao Ma, Xiyao Ye, Zack Han, Xu Yin, Jun Ji, Heng Shen, Wei Fan, Xing Yao, Benjamin Guo, Chenlei |
| contents | We introduce WildAGTEval, a benchmark designed to evaluate large language model (LLM) agents' function-calling capabilities under realistic API complexity. Unlike prior work that assumes an idealized API system and disregards real-world factors such as noisy API outputs, WildAGTEval accounts for two dimensions of real-world complexity: 1. API specification, which includes detailed documentation and usage constraints, and 2. API execution, which captures runtime challenges. Consequently, WildAGTEval offers (i) an API system encompassing 60 distinct complexity scenarios that can be composed into approximately 32K test configurations, and (ii) user-agent interactions for evaluating LLM agents on these scenarios. Using WildAGTEval, we systematically assess several advanced LLMs and observe that most scenarios are challenging, with irrelevant information complexity posing the greatest difficulty and reducing the performance of strong LLMs by 27.3%. Furthermore, our qualitative analysis reveals that LLMs occasionally distort user intent merely to claim task completion, critically affecting user satisfaction. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_00268 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity Kim, Doyoung Ren, Zhiwei Hao, Jie Sun, Zhongkai Wang, Lichao Ma, Xiyao Ye, Zack Han, Xu Yin, Jun Ji, Heng Shen, Wei Fan, Xing Yao, Benjamin Guo, Chenlei Computation and Language Artificial Intelligence We introduce WildAGTEval, a benchmark designed to evaluate large language model (LLM) agents' function-calling capabilities under realistic API complexity. Unlike prior work that assumes an idealized API system and disregards real-world factors such as noisy API outputs, WildAGTEval accounts for two dimensions of real-world complexity: 1. API specification, which includes detailed documentation and usage constraints, and 2. API execution, which captures runtime challenges. Consequently, WildAGTEval offers (i) an API system encompassing 60 distinct complexity scenarios that can be composed into approximately 32K test configurations, and (ii) user-agent interactions for evaluating LLM agents on these scenarios. Using WildAGTEval, we systematically assess several advanced LLMs and observe that most scenarios are challenging, with irrelevant information complexity posing the greatest difficulty and reducing the performance of strong LLMs by 27.3%. Furthermore, our qualitative analysis reveals that LLMs occasionally distort user intent merely to claim task completion, critically affecting user satisfaction. |
| title | Beyond Perfect APIs: A Comprehensive Evaluation of LLM Agents Under Real-World API Complexity |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2601.00268 |