Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents

• URL: http://arxiv.org/abs/2606.19704v1
• Date: Thu, 18 Jun 2026 02:02:46 GMT
• Title: Beyond Static Leaderboards: Predictive Validity for the Evaluation of LLM Agents
• Abstract summary: This paper aggregates the largest coordinated deep-dive of one MCP-based industrial-agent benchmark to date.<n>We argue that aggregate-score leaderboards systematically underspecify deployed-agent evaluation.
• Score: 19.290397279235084
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Agent benchmarks are growing fast, but no single benchmark touches more than four or five of the dimensions that deployment exposes. This paper aggregates the largest coordinated deep-dive of one MCP-based industrial-agent benchmark to date: fourteen parallel implementation studies covering new asset classes (including a multi-modal visual extension), alternative orchestrations, retrieval strategies, reasoning modes, infrastructure optimizations, and evaluation-methodology probes. Consolidating those studies with seven prior agent benchmarks, we argue that aggregate-score leaderboards systematically underspecify deployed-agent evaluation. Rankings derived from aggregate scores do not transfer to out-of-distribution settings; recent public-to-hidden competition retrospectives provide direct empirical evidence of this rank instability. We propose ranking configurations by predictive validity, the correlation between in-sample and out-of-sample rank, rather than in-sample mean, and report a twelve-tier measurement apparatus that exposes the deployment-relevant dimensions HELM and its agent-era successors collapse. The position is operationalized through three falsifiable out-of-distribution criteria with explicit thresholds; existing evidence partly supports it but is too thin to confirm. We close with a pre-registered pilot design and a field-level vision for what the next generation of agentic benchmarks should report.

Related papers

• The Double Measurement Confound in Agent Benchmarks: De-Scaffolding, Ground-Truth Scoring, and Reliability Beyond the Mean [2.822987192839875]
  Agent benchmarks are increasingly used to compare large language models (LLMs) and guide deployment decisions.<n>We identify a double measurement confound: execution-critical decisions are performed by a fixed scaffold instead of the model, while the scorer evaluates outputs using criteria that may not reflect task correctness.<n>We unify these issues within a measurement-theoretic framework that characterizes when benchmark scores can be interpreted as evidence of model capability.
  arXiv  Detail & Related papers  (2026-09-06T21:23:11Z)
• MedSegBenchmarker: A Raw-Count-First Framework for Controlled 2D Medical Image Segmentation Benchmarks [1.0230975056328007]
  We present MEDSEGBENCHMARKER (MSB), a configuration-driven framework for controlled benchmarking of 2D datasets.<n>MSB exports sample- and class-level pixel counts and predictions together with the evaluation context.<n>We demonstrate MSB in a case study involving three heterogeneous 2D datasets and multiple vision models evaluated at 256- and 512-pixel input resolutions.
  arXiv  Detail & Related papers  (2026-08-30T09:20:38Z)
• Position: Evaluation Scores Are Perishable Knowledge Claims [0.0]
  We argue that evaluation scores should be treated as claims with three properties: formality, scope, and validity windows.<n>We illustrate the cost of mean aggregation on the public HELM leaderboard.
  arXiv  Detail & Related papers  (2026-07-28T18:50:39Z)
• PACE: A Proxy for Agentic Capability Evaluation [60.3743414796937]
  We introduce PACE, a framework that constructs proxy benchmarks by selecting instances from existing non-agentic evaluations.<n>Experiments across 14 models, 4 agentic benchmarks, and 19 non-agentic benchmarks show that PACE-Bench predicts agentic scores with leave-one-out cross-validation (LOOCV) mean absolute error (MAE) under 4%, Spearman correlation above 0.80, and pairwise model-ranking accuracy around 85%, all at much less than 1% of the full agentic evaluation cost.
  arXiv  Detail & Related papers  (2026-07-02T10:59:03Z)
• AgentBeats: Agentifying Agent Assessment for Openness, Standardization, and Reproducibility [104.46861849039357]
  Agent systems are advancing quickly across domains, but their evaluation remains fragmented.<n>The root problem is the lack of an open, agent-agnostic assessment interface.<n>We advocate Agentified Agent Assessment (AAA), where evaluation is performed by judge agents and all participants interact through standardized protocols.
  arXiv  Detail & Related papers  (2026-06-11T17:23:54Z)
• A Unified Framework for the Evaluation of LLM Agentic Capabilities [36.43241368835721]
  We present a unified framework for the fair evaluation of LLM agentic capabilities.<n> Driven by a unified configuration system, the framework integrates diverse benchmarks into a standardized instruction--tool--environment format.<n>We conduct a large-scale empirical analysis over 400K rollouts and 5B tokens on 15 models.
  arXiv  Detail & Related papers  (2026-05-27T03:20:45Z)
• AI Cartography: Mapping the Latent Landscape of AI Benchmark Ecosystems [20.174278188532096]
  We introduce a framework for measuring the latent landscape in AI benchmark ecosystems.<n>Applying Confirmatory Factor Analysis (CFA) and Generalizability Theory to 4,000+ models from the Open LLM Leaderboard, we decompose sources of ranking variance.<n>We provide actionable diagnostics to determine how benchmark rankings can be trusted and how benchmark design can be improved.
  arXiv  Detail & Related papers  (2026-05-24T21:59:08Z)
• Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents? [61.49434544687523]
  We introduce REFLECT, a meta-evaluation benchmark targeting fine-grained failure detection in agentic environments.<n> REFLECT defines a detailed taxonomy of process- and outcome-level failure modes, instantiated by performing controlled and localized interventions.<n>Our experiments show that even the best-performing models achieve overall accuracies below 55% across reasoning, tool-use, and report-quality failures.
  arXiv  Detail & Related papers  (2026-05-18T23:55:08Z)
• RubricEM: Meta-RL with Rubric-guided Policy Decomposition beyond Verifiable Rewards [76.17893114021757]
  Deep research systems that plan, search, evaluate evidence, and synthesize long-form reports lack ground-truth answers and span many tool-augmented decisions.<n>In this work, we argue that rubrics should serve not merely as final-answer evaluators, but as the shared interface that structures policy execution, judge feedback, and agent memory.<n>We introduceEM, a rubric-guided reinforcement learning framework that combines stagewise policy decomposition with reflection-based meta-policy evolution.
  arXiv  Detail & Related papers  (2026-05-11T17:40:38Z)
• Deployment-Relevant Alignment Cannot Be Inferred from Model-Level Evaluation Alone [11.663456969895462]
  Alignment evaluation in machine learning has largely become evaluation of models.<n>This paper argues that deployment-relevant alignment cannot be inferred from model-level evaluation alone.
  arXiv  Detail & Related papers  (2026-05-06T03:28:30Z)
• Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems [55.04168927059962]
  Reasoning-intensive retrieval aims to surface evidence that supports downstream reasoning rather than merely matching topical similarity.<n>We introduce BRIGHT-Pro, an expert-annotated benchmark that expands each query with multi-aspect gold evidence.<n>We also construct RTriever- Synth, an aspect-decomposed synthetic corpus that generates complementary positives and positive-conditioned hard negatives.
  arXiv  Detail & Related papers  (2026-05-05T17:42:50Z)
• RPA-Check: A Multi-Stage Automated Framework for Evaluating Dynamic LLM-based Role-Playing Agents [4.07447364754644]
  This paper introduces RPA-Check, a multi-stage automated evaluation framework designed to objectively assess the performance of LLM-based RPAs.<n>We validate this framework by applying it to LLM Court, a serious game for forensic training involving several quantized local models.
  arXiv  Detail & Related papers  (2026-04-13T16:08:03Z)
• Efficient Benchmarking of AI Agents [0.0]
  We study whether small task subsets can preserve agent rankings at substantially lower cost.<n>We find that absolute score prediction degrades under scaffold-driven distribution shift.<n>We propose an optimization-free protocol: evaluate new agents only on tasks with intermediate historical pass rates.
  arXiv  Detail & Related papers  (2026-03-24T22:17:11Z)
• Benchmark^2: Systematic Evaluation of LLM Benchmarks [66.2731798872668]
  We propose Benchmark2, a comprehensive framework comprising three complementary metrics.<n>We conduct experiments across 15 benchmarks spanning mathematics, reasoning, and knowledge domains.<n>Our analysis reveals significant quality variations among existing benchmarks and demonstrates that selective benchmark construction can achieve comparable evaluation performance.
  arXiv  Detail & Related papers  (2026-01-07T14:59:03Z)
• Structured Prompting Enables More Robust Evaluation of Language Models [38.53918044830268]
  We present a DSPy+HELM framework that introduces structured prompting methods which elicit reasoning.<n>We find that without structured prompting, HELM underestimates LM performance (by 4% average) and performance estimates vary more across benchmarks.<n>This is the first benchmarking study to systematically integrate structured prompting into an established evaluation framework.
  arXiv  Detail & Related papers  (2025-11-25T20:37:59Z)

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.