SCAN: Structured Capability Assessment and Navigation for LLMs
Not on the map. Why?
Abstract (arXiv)
Evaluating Large Language Models (LLMs) has become increasingly important, with automatic evaluation benchmarks gaining prominence as alternatives to human evaluation. While existing research has focused on approximating model rankings, such benchmarks fail to provide users and developers with a comprehensive and fine-grained understanding of a specific model's capabilities. To fill this gap, we propose SCAN (Structured Capability Assessment and Navigation), a practical framework that enables detailed characterization of LLM capabilities through comprehensive and fine-grained evaluation. SCAN incorporates four key components: (1) TaxBuilder, which extracts capability-indicating tags from extensive queries to construct a hierarchical taxonomy automatically; (2) RealMix, a query synthesis and filtering mechanism that ensures sufficient evaluation data for each capability tag; (3) a suite of visualization and analysis tools that facilitate efficient navigation and analysis of model capabilities; and (4) a PC2-based (Pre-Comparison-derived Criteria) LLM-as-a-Judge approach that achieves significantly higher accuracy compared to classic LLM-as-a-Judge method. Using SCAN, we conduct a comprehensive evaluation of 21 mainstream LLMs. Our detailed analysis of the GPT-OSS family reveals substantial performance variations, even within sub-capabilities belonging to the same category of capability. This finding highlights the importance of fine-grained evaluation in accurately understanding LLM behavior. Project homepage and resources are available at hrefhttps://github.com/liudan193/SCANhttps://github.com/liudan193/SCAN.
Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 27 of its 59 references that are in the map. First posted on arXiv 10 May 2025.
Builds on
27 of 59 references.
References · 27
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
- 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,626
- 2022Self-Instruct: Aligning Language Models with Self-Generated Instructionscited by 2,724
- 2024Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preferencecited by 1,184
- 2024Length-Controlled AlpacaEval: A Simple Way to Debias Automatic Evaluatorscited by 897
- 2023Prometheus: Inducing Fine-grained Evaluation Capability in Language Modelscited by 575
- 2024From Crowdsourced Data to High-Quality Benchmarks: Arena-Hard and BenchBuilder Pipelinecited by 500
- 2024From Generation to Judgment: Opportunities and Challenges of LLM-as-a-judgecited by 494
- 2024LLMs-as-Judges: A Comprehensive Survey on LLM-based Evaluation Methodscited by 465
- 2023Perception Test: A Diagnostic Benchmark for Multimodal Video Modelscited by 314
- 2024WildBench: Benchmarking LLMs with Challenging Tasks from Real Users in the Wildcited by 218
- 2024RegMix: Data Mixture as Regression for Language Model Pre-trainingcited by 177
15 more
- 2023Platypus: Quick, Cheap, and Powerful Refinement of LLMscited by 150
- 2024SafetyPrompts: a Systematic Review of Open Datasets for Evaluating and Improving Large Language Model Safetycited by 77
- 2024MixEval: Deriving Wisdom of the Crowd from LLM Benchmark Mixturescited by 74
- 2024Lynx: An Open Source Hallucination Evaluation Modelcited by 57
- 2024FairMedFM: Fairness Benchmarking for Medical Imaging Foundation Modelscited by 47
- 2024Auto-Arena: Automating LLM Evaluations with Agent Peer Battles and Committee Discussionscited by 42
- 2024CompassJudger-1: All-in-one Judge Model Helps Model Evaluation and Evolutioncited by 42
- 2024HalluDial: A Large-Scale Benchmark for Automatic Dialogue-Level Hallucination Evaluationcited by 27
- 2025EvalTree: Profiling Language Model Weaknesses via Hierarchical Capability Treescited by 27
- 2025DOVE: A Large-Scale Multi-Dimensional Predictions Dataset Towards Meaningful LLM Evaluationcited by 21
- 2025QuaDMix: Quality-Diversity Balanced Data Selection for Efficient LLM Pretrainingcited by 16
- 2024Hallu-PI: Evaluating Hallucination in Multi-modal Large Language Models within Perturbed Inputscited by 14
- 2024FB-Bench: A Fine-Grained Multi-Task Benchmark for Evaluating LLMs' Responsiveness to Human Feedbackcited by 11
- 2024Self-Judge: Selective Instruction Following with Alignment Self-Evaluationcited by 7
- 2024C2LEVA: Toward Comprehensive and Contamination-Free Language Model Evaluationcited by 2
Built on by
Citing papers per year: 2026: 2.
2 papers.