# GroundEval: A Deterministic Replacement for LLM-as-Judge in Stateful Agent Evaluation

arXiv: https://arxiv.org/abs/2606.22737 · HTML: https://mlsanity.com/paper/2606.22737/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 22 of its 23 references that are in the map.
- First posted on arXiv 22 Jun 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 22 of 22)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [Evaluating Very Long-Term Conversational Memory of LLM Agents](https://mlsanity.com/paper/2402.17753/) (2024)
- [LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memory](https://mlsanity.com/paper/2410.10813/) (2024)
- [Measuring Faithfulness in Chain-of-Thought Reasoning](https://mlsanity.com/paper/2307.13702/) (2023)
- [Ragas: Automated Evaluation of Retrieval Augmented Generation](https://mlsanity.com/paper/2309.15217/) (2023)
- [Justice or Prejudice? Quantifying Biases in LLM-as-a-Judge](https://mlsanity.com/paper/2410.02736/) (2024)
- [AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents](https://mlsanity.com/paper/2401.13178/) (2024)
- [Faithful Chain-of-Thought Reasoning](https://mlsanity.com/paper/2301.13379/) (2023)
- [RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models](https://mlsanity.com/paper/2401.00396/) (2023)
- [Judging the Judges: A Systematic Study of Position Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2406.07791/) (2024)
- [ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2311.09476/) (2023)
- [Self-Preference Bias in LLM-as-a-Judge](https://mlsanity.com/paper/2410.21819/) (2024)
- [Measuring Attribution in Natural Language Generation Models](https://mlsanity.com/paper/2112.12870/) (2021)
- [AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories](https://mlsanity.com/paper/2504.08942/) (2025)
- [TRAJECT-Bench:A Trajectory-Aware Benchmark for Evaluating Agentic Tool Use](https://mlsanity.com/paper/2510.04550/) (2025)
- [LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleagues](https://mlsanity.com/paper/2605.12493/) (2026)
- [OrgForge-IT: A Verifiable Synthetic Benchmark for LLM-Based Insider Threat Detection](https://mlsanity.com/paper/2603.22499/) (2026)
- [Structured Belief State and the First Precision-Aware Benchmark for LLM Memory Retrieval](https://mlsanity.com/paper/2605.11325/) (2026)
- [OrgForge: A Multi-Agent Simulation Framework for Verifiable Synthetic Corporate Corpora](https://mlsanity.com/paper/2603.14997/) (2026)

## Built on by (most cited first, 6 of 6)

- [When May an Agent Stop? Evidence-Carrying Termination for Tool-Using LLMs](https://mlsanity.com/paper/2608.23623/) (2026)
- [From Checker to Forecaster: Code-Owned Evaluation of Model-Generated Strategic Routes Under Delayed Ground Truth](https://mlsanity.com/paper/2607.10972/) (2026)
- [Win by Silence: Deletion Non-Monotonicity, Autonomous Exploitation, and Typed-State Gating in LLM Plan Evaluation](https://mlsanity.com/paper/2607.12986/) (2026)
- [Continuous Process-Level Evaluation for Evolving Enterprise AI Agent Skills](https://mlsanity.com/paper/2610.01833/) (2026)
- [Open-Endedness Bench: Measuring Epistemic Process from Agent Records](https://mlsanity.com/paper/2610.02588/) (2026)
- [One Step at a Time: Trading LLM Autonomy for Process Predictability](https://mlsanity.com/paper/2610.07817/) (2026)

## Abstract (arXiv)

Before letting an agent operate over real context, can you prove it used the right evidence? GroundEval turns that question into a deterministic test of what the agent searched, fetched, cited, and was permitted to access. In one case study, two frontier LLM judges scored a plausible agent response 0.85 and higher. But the trace told a different story: the agent had never retrieved the artifact its answer depended on, yielding a GroundEval score of 0.000. We introduce GroundEval, a judge-free framework for evaluating agents against grounded, time-bounded, and access-controlled evidence. GroundEval uses a domain configuration to generate questions, lets the agent choose how to answer, and then scores both the final answer and the recorded trajectory that produced it. The benchmark targets three failures that LLM-as-judge evaluation struggles to detect: whether an agent checked before claiming absence, reasoned only from evidence available to the actor at the relevant time, and used the correct causal mechanism rather than a plausible one. These correspond to three tracks: Silence, Perspective, and Counterfactual. GroundEval exposes when plausible answers rest on invalid evidence paths, and produces structured per-question diagnostics that pair tool activity with the agent's turn-level narration, making each score inspectable rather than merely reported. Our case studies suggest this failure mode is common rather than exceptional, one that final-answer and judge-based evaluation cannot detect by construction.