KnowledgeBerg: Evaluating Systematic Knowledge Coverage and Compositional Reasoning in Large Language Models

TL;DR

KnowledgeBerg is a comprehensive benchmark testing large language models' ability to systematically cover knowledge domains and perform compositional reasoning, revealing significant limitations across multiple models and languages.

Contribution

The paper introduces KnowledgeBerg, a new benchmark with 4,800 questions across diverse domains and languages, to evaluate LLMs' knowledge coverage and reasoning capabilities.

Findings

Open-source LLMs perform poorly on universe enumeration and reasoning tasks.

Test-time augmentation improves model performance by up to 4.35 points.

Failures are due to missing knowledge, lack of awareness, and incorrect reasoning execution.

Abstract

Many real-world questions appear deceptively simple yet implicitly demand two capabilities: (i) systematic coverage of a bounded knowledge universe and (ii) compositional set-based reasoning over that universe, a phenomenon we term "the tip of the iceberg." We formalize this challenge through two orthogonal dimensions: knowledge width, the cardinality of the required universe, and reasoning depth, the number of compositional set operations. We introduce KnowledgeBerg, a benchmark of 4,800 multiple-choice questions derived from 1,183 enumeration seeds spanning 10 domains and 17 languages, with universes grounded in authoritative sources to ensure reproducibility. Representative open-source LLMs demonstrate severe limitations, achieving only 5.26-36.88 F1 on universe enumeration and 16.00-44.19 accuracy on knowledge-grounded reasoning. Diagnostic analyses reveal three stages of failure: completeness, or missing knowledge; awareness, or failure to identify requirements; and application, or incorrect reasoning execution. This pattern persists across languages and model scales. Although test-time compute and retrieval augmentation yield measurable gains -- up to 4.35 and 3.78 points, respectively -- substantial gaps remain, exposing limitations in how current LLMs organize structured knowledge and execute compositional reasoning over bounded domains. The dataset is available at https://huggingface.co/datasets/2npc/KnowledgeBerg