analiza aiqbench
CorporateBench addresses a critical gap in LLM evaluation by introducing a large-scale, human-validated benchmark that simulates realistic corporate communication networks with over 230,000 documents. The study reveals that current models struggle significantly as input size approaches the scale of actual enterprise environments, highlighting limitations in handling temporal knowledge consistency and complex information extraction. This resource provides developers with a necessary metric to assess reasoning capabilities beyond simple synthetic datasets.
CorporateBench: Large-Scale Q&A Benchmarking with Temporal Knowledge Bases
arXiv:2608.27391v1 Announce Type: new Abstract: LLMs are increasingly able to answer complex questions about enterprise-scale document collections. But evaluation is hard: companies don't want to share internal communications, and synthetic datasets have been overly simple. We present CorporateBench (CB), a human-validated multi-task Q&A benchmark whose scale approaches the conditions LLMs encounter in corporate communication networks, with evaluation corpora surpassing 230,000 documents. CB evaluates LLMs across two dimensions (information extraction and knowledge base querying) through four synthetically generated firms ranging from 12 to 10,000 employees. Each corpus is sampled from a temporally evolving knowledge base describing a consistent world, guaranteeing cross-document logical consistency even across hundreds of thousands of documents. We evaluate five LLMs on CB, revealing increasingly poor performance as input size approaches realistic scales. CB provides LLM developers a metric for corporate communication reasoning, filling a crucial gap in the benchmarking ecosystem.
Abstract:LLMs are increasingly able to answer complex questions about enterprise-scale document collections. But evaluation is hard: companies don't want to share internal communications, and synthetic datasets have been overly simple. We present CorporateBench (CB), a human-validated multi-task Q&A benchmark whose scale approaches the conditions LLMs encounter in corporate communication networks, with evaluation corpora surpassing 230,000 documents. CB evaluates LLMs across two dimensions (information extraction and knowledge base querying) through four synthetically generated firms ranging from 12 to 10,000 employees. Each corpus is sampled from a temporally evolving knowledge base describing a consistent world, guaranteeing cross-document logical consistency even across hundreds of thousands of documents. We evaluate five LLMs on CB, revealing increasingly poor performance as input size approaches realistic scales. CB provides LLM developers a metric for corporate communication reasoning, filling a crucial gap in the benchmarking ecosystem.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR); Machine Learning (cs.LG)
arXiv:2608.27391 [cs.AI]
(or arXiv:2608.27391v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2608.27391 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 242 cuvinte
Subiecte conexe
[AINews] Opus 5.5 is good at explainer videos
Latent Space · 29 sept. 2026
[AINews] The Future of Latent Space
Latent Space · 25 sept. 2026
Import AI 476: Creativity and RSI; worse-than-China open weight models; AI goes to space
Import AI · 12 oct. 2026
Links #5: 2026/07
LessWrong · 31 iul. 2026