Toward a unified framework for data-efficient evaluation of large language models

• URL: http://arxiv.org/abs/2510.04051v1
• Date: Sun, 05 Oct 2025 06:13:50 GMT
• Title: Toward a unified framework for data-efficient evaluation of large language models
• Authors: Lele Liao, Qile Zhang, Ruofan Wu, Guanhua Fang,
• Abstract summary: LEGO-IRT is a unified and flexible framework for data-efficient large language models evaluation.<n>It supports both binary and continuous evaluation metrics.<n>We show that LEGO-IRT achieves stable capability estimates using just $3%$ of the total evaluation items.
• Score: 12.922829524961813
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Evaluating large language models (LLMs) on comprehensive benchmarks is a cornerstone of their development, yet it's often computationally and financially prohibitive. While Item Response Theory (IRT) offers a promising path toward data-efficient evaluation by disentangling model capability from item difficulty, existing IRT-based methods are hampered by significant limitations. They are typically restricted to binary correctness metrics, failing to natively handle the continuous scores used in generative tasks, and they operate on single benchmarks, ignoring valuable structural knowledge like correlations across different metrics or benchmarks. To overcome these challenges, we introduce LEGO-IRT, a unified and flexible framework for data-efficient LLM evaluation. LEGO-IRT's novel design natively supports both binary and continuous evaluation metrics. Moreover, it introduces a factorized architecture to explicitly model and leverage structural knowledge, decomposing model ability estimates into a general component and structure-specific (e.g., per-metric or per-benchmark) components. Through extensive experiments involving $70$ LLMs across $5$ benchmarks, we show that LEGO-IRT achieves stable capability estimates using just $3\%$ of the total evaluation items. We demonstrate that incorporating structural knowledge reduces estimation error by up to $10\%$ and reveal that the latent abilities estimated by our framework may align more closely with human preferences.

Related papers

• Beyond Accuracy: Characterizing Code Comprehension Capabilities in (Large) Language Models [4.841487377596519]
  This paper investigates whether Large Language Models' code-comprehension performance aligns with traditional human-centric software metrics.<n>We introduce a diagnostic framework that reframes code understanding as a binary input-output consistency task.
  arXiv  Detail & Related papers  (2026-01-19T10:58:24Z)
• Table-BiEval: A Self-Supervised, Dual-Track Framework for Decoupling Structure and Content in LLM Evaluation [11.450834626205676]
  Table-BiEval is a novel approach based on a human-free, self-supervised evaluation framework.<n>It calculates Content Semantic Accuracy and Normalized Tree Edit Distance to decouple structure from content.<n>Results reveal substantial variability, highlighting that mid-sized models can surprisingly outperform larger counterparts in structural efficiency.
  arXiv  Detail & Related papers  (2026-01-09T07:38:27Z)
• LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering [90.84806758077536]
  We introduce textbfLoCoBench-Agent, a comprehensive evaluation framework specifically designed to assess large language models (LLMs) agents in realistic, long-context software engineering.<n>Our framework extends LoCoBench's 8,000 scenarios into interactive agent environments, enabling systematic evaluation of multi-turn conversations.<n>Our framework provides agents with 8 specialized tools (file operations, search, code analysis) and evaluates them across context lengths ranging from 10K to 1M tokens.
  arXiv  Detail & Related papers  (2025-11-17T23:57:24Z)
• LTD-Bench: Evaluating Large Language Models by Letting Them Draw [57.237152905238084]
  LTD-Bench is a breakthrough benchmark for large language models (LLMs)<n>It transforms LLM evaluation from abstract scores to directly observable visual outputs by requiring models to generate drawings through dot matrices or executable code.<n> LTD-Bench's visual outputs enable powerful diagnostic analysis, offering a potential approach to investigate model similarity.
  arXiv  Detail & Related papers  (2025-11-04T08:11:23Z)
• StoryBench: A Dynamic Benchmark for Evaluating Long-Term Memory with Multi Turns [7.60350050736492]
  Long-term memory is essential for large language models to achieve autonomous intelligence.<n>Existing benchmarks face challenges in evaluating knowledge retention and dynamic sequential reasoning.<n>We propose a novel benchmark framework based on interactive fiction games.
  arXiv  Detail & Related papers  (2025-06-16T10:54:31Z)
• DSR-Bench: Evaluating the Structural Reasoning Abilities of LLMs via Data Structures [20.596558700597644]
  Large language models (LLMs) are increasingly deployed for real-world tasks that fundamentally involve data manipulation.<n>A core requirement is the ability to perform structural reasoning--that is, to understand and reason about data relationships.<n>We introduce DSR-Bench, a novel benchmark evaluating LLMs' structural reasoning capabilities through data structures.
  arXiv  Detail & Related papers  (2025-05-29T23:24:53Z)
• Lost in Benchmarks? Rethinking Large Language Model Benchmarking with Item Response Theory [44.886213907135435]
  We propose Pseudo-Siamese Network for Item Response Theory (PSN-IRT) for accurate and reliable estimations of item characteristics and model abilities.<n>PSN-IRT can be utilized for accurate and reliable estimations of item characteristics and model abilities.
  arXiv  Detail & Related papers  (2025-05-21T03:24:11Z)
• Model Utility Law: Evaluating LLMs beyond Performance through Mechanism Interpretable Metric [99.56567010306807]
  Large Language Models (LLMs) have become indispensable across academia, industry, and daily applications.<n>One core challenge of evaluation in the large language model (LLM) era is the generalization issue.<n>We propose Model Utilization Index (MUI), a mechanism interpretability enhanced metric that complements traditional performance scores.
  arXiv  Detail & Related papers  (2025-04-10T04:09:47Z)
• StructTest: Benchmarking LLMs' Reasoning through Compositional Structured Outputs [78.84060166851805]
  StructTest is a novel benchmark that evaluates large language models (LLMs) on their ability to follow compositional instructions and generate structured outputs.<n> Assessments are conducted deterministically using a rule-based evaluator, which can be easily extended to new tasks and datasets.<n>We demonstrate that StructTest remains challenging even for top-performing models like Deepseek-V3/R1 and GPT-4o.
  arXiv  Detail & Related papers  (2024-12-23T22:08:40Z)
• The Vulnerability of Language Model Benchmarks: Do They Accurately Reflect True LLM Performance? [1.3810901729134184]
  Large Language Models (LLMs) excel at standardized tests while failing to demonstrate genuine language understanding and adaptability.<n>Our systematic analysis of NLP evaluation frameworks reveals pervasive vulnerabilities across the evaluation spectrum.<n>We lay the groundwork for new evaluation methods that resist manipulation, minimize data contamination, and assess domain-specific tasks.
  arXiv  Detail & Related papers  (2024-12-02T20:49:21Z)
• Benchmarks as Microscopes: A Call for Model Metrology [76.64402390208576]
  Modern language models (LMs) pose a new challenge in capability assessment. To be confident in our metrics, we need a new discipline of model metrology.
  arXiv  Detail & Related papers  (2024-07-22T17:52:12Z)
• KoLA: Carefully Benchmarking World Knowledge of Large Language Models [87.96683299084788]
  We construct a Knowledge-oriented LLM Assessment benchmark (KoLA) We mimic human cognition to form a four-level taxonomy of knowledge-related abilities, covering $19$ tasks. We use both Wikipedia, a corpus prevalently pre-trained by LLMs, along with continuously collected emerging corpora, to evaluate the capacity to handle unseen data and evolving knowledge.
  arXiv  Detail & Related papers  (2023-06-15T17:20:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.