# Time to REFLECT: Can We Trust LLM Judges for Evidence-based Research Agents?

arXiv: https://arxiv.org/abs/2605.19196 · HTML: https://mlsanity.com/paper/2605.19196/

## Key facts

- Cited by 10 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 55 of its 65 references that are in the map.
- First posted on arXiv 18 May 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, 30 of 55)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Direct Preference Optimization: Your Language Model is Secretly a Reward Model](https://mlsanity.com/paper/2305.18290/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Gemma 3 Technical Report](https://mlsanity.com/paper/2503.19786/) (2025)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://mlsanity.com/paper/2305.14251/) (2023)
- [AlpacaFarm: A Simulation Framework for Methods that Learn from Human Feedback](https://mlsanity.com/paper/2305.14387/) (2023)
- [Enabling Large Language Models to Generate Text with Citations](https://mlsanity.com/paper/2305.14627/) (2023)
- [Ragas: Automated Evaluation of Retrieval Augmented Generation](https://mlsanity.com/paper/2309.15217/) (2023)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [Benchmarking Large Language Models in Retrieval-Augmented Generation](https://mlsanity.com/paper/2309.01431/) (2023)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)
- [ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities](https://mlsanity.com/paper/2408.04682/) (2024)
- [RAGTruth: A Hallucination Corpus for Developing Trustworthy Retrieval-Augmented Language Models](https://mlsanity.com/paper/2401.00396/) (2023)
- [ARES: An Automated Evaluation Framework for Retrieval-Augmented Generation Systems](https://mlsanity.com/paper/2311.09476/) (2023)
- [DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents](https://mlsanity.com/paper/2506.11763/) (2025)
- [Agent-as-a-Judge: Evaluate Agents with Agents](https://mlsanity.com/paper/2410.10934/) (2024)
- [MetaTool Benchmark for Large Language Models: Deciding Whether to Use Tools and Which to Use](https://mlsanity.com/paper/2310.03128/) (2023)
- [Long-form factuality in large language models](https://mlsanity.com/paper/2403.18802/) (2024)
- [RewardBench 2: Advancing Reward Model Evaluation](https://mlsanity.com/paper/2506.01937/) (2025)
- [Where LLM Agents Fail and How They can Learn From Failures](https://mlsanity.com/paper/2509.25370/) (2025)
- [AgentRewardBench: Evaluating Automatic Evaluations of Web Agent Trajectories](https://mlsanity.com/paper/2504.08942/) (2025)
- [FaithEval: Can Your Language Model Stay Faithful to Context, Even If "The Moon is Made of Marshmallows"](https://mlsanity.com/paper/2410.03727/) (2024)
- [ResearchRubrics: A Benchmark of Prompts and Rubrics For Evaluating Deep Research Agents](https://mlsanity.com/paper/2511.07685/) (2025)

## Built on by (most cited first, 10 of 10)

- [Autonomous Research Agents: A Survey of AI Scientists and the Verification Gap](https://mlsanity.com/paper/2608.05179/) (2026)
- [Open Evaluation Agent: Efficient and Promptable Evaluation of Visual Generative Models](https://mlsanity.com/paper/2608.09666/) (2026)
- [trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories](https://mlsanity.com/paper/2609.00038/) (2026)
- [Workflow Closure Is Not Scientific Closure in Auto-Research Systems](https://mlsanity.com/paper/2605.26200/) (2026)
- [Eval-Pair Matrix: Answer-Paired Meta-Evaluation of LLM Judges for Grounded RAG](https://mlsanity.com/paper/2607.10626/) (2026)
- [Forward-Deployed Full-Stack Engineering for Autonomous Cloud MLOps](https://mlsanity.com/paper/2608.29615/) (2026)
- [RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation](https://mlsanity.com/paper/2609.16900/) (2026)
- [The Hard Part Comes After Search: Benchmarking Web Agents on Synthesizing, Organizing, and Displaying Knowledge](https://mlsanity.com/paper/2609.30604/) (2026)
- [Learning Strategies to Break Judges](https://mlsanity.com/paper/2609.33773/) (2026)
- [PADMÉ: Preference Alignment Data Synthesis for Meta-Evaluation of LM Agent Evaluators](https://mlsanity.com/paper/2609.36086/) (2026)

## Abstract (arXiv)

Deep research agents increasingly automate complex information-seeking tasks, producing evidence-grounded reports via multi-step reasoning, tool use, and synthesis. Their growing role demands scalable, reliable evaluation, positioning LLM-as-judge as a supervision paradigm for assessing factual accuracy, evidence use, and reasoning quality. Yet the reliability of these judges for deep research agents remains poorly understood, posing a critical meta-evaluation problem: before deploying LLM judges to supervise research agents, we must first evaluate the judges themselves. Existing meta-evaluations fall short in two ways: (1) reliance on coarse, subjective human-preference agreement; (2) focus on instruction-following or verifiable tasks, leaving open-ended agent executions unexplored. To address these gaps, we introduce REFLECT (REliable Fine-grained LLM judge Evaluation via Controlled inTervention), a meta-evaluation benchmark targeting fine-grained failure detection in agentic environments. REFLECT defines a detailed taxonomy of process- and outcome-level failure modes, instantiated by performing controlled and localized interventions on quality-screened agent execution traces. This yields verifiable, comprehensive, and fine-grained instances for validating the judge models. Our experiments show that current LLM judges remain unreliable: even the best-performing models achieve overall accuracies below 55% across reasoning, tool-use, and report-quality failures, with especially poor performance on evidence verification. Together, our taxonomy and findings expose systematic judge limitations, reveal tradeoffs in cost and reliability, and offer actionable guidance for building more reliable evaluation pipelines for deep research agents.