# trajectory-judge: What Outcome-Only LLM Judges Miss on Agent Trajectories

arXiv: https://arxiv.org/abs/2609.00038 · HTML: https://mlsanity.com/paper/2609.00038/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in General LLM agent frameworks and benchmarks, one of 16 directions in LLM agents and tool use.
- Builds on 51 of its 60 references that are in the map.
- First posted on arXiv 29 Aug 2026.

## Placed in

- LLM agents and tool use / [General LLM agent frameworks and benchmarks](https://mlsanity.com/agents/general-llm-agent-frameworks-and-benchmarks/)

## Builds on (most cited first, 30 of 51)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Self-Consistency Improves Chain of Thought Reasoning in Language Models](https://mlsanity.com/paper/2203.11171/) (2022)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [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)
- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [Language Models (Mostly) Know What They Know](https://mlsanity.com/paper/2207.05221/) (2022)
- [A Survey on LLM-as-a-Judge](https://mlsanity.com/paper/2411.15594/) (2024)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [AgentBench: Evaluating LLMs as Agents](https://mlsanity.com/paper/2308.03688/) (2023)
- [Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Prompting](https://mlsanity.com/paper/2305.04388/) (2023)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Can LLMs Express Their Uncertainty? An Empirical Evaluation of Confidence Elicitation in LLMs](https://mlsanity.com/paper/2306.13063/) (2023)
- [Math-Shepherd: Verify and Reinforce LLMs Step-by-step without Human Annotations](https://mlsanity.com/paper/2312.08935/) (2023)
- [Solving math word problems with process- and outcome-based feedback](https://mlsanity.com/paper/2211.14275/) (2022)
- [Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback](https://mlsanity.com/paper/2305.14975/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [τ2-Bench: Evaluating Conversational Agents in a Dual-Control Environment](https://mlsanity.com/paper/2506.07982/) (2025)
- [Why Do Multi-Agent LLM Systems Fail?](https://mlsanity.com/paper/2503.13657/) (2025)
- [The Lessons of Developing Process Reward Models in Mathematical Reasoning](https://mlsanity.com/paper/2501.07301/) (2025)
- [Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation](https://mlsanity.com/paper/2503.11926/) (2025)
- [AgentBoard: An Analytical Evaluation Board of Multi-turn LLM Agents](https://mlsanity.com/paper/2401.13178/) (2024)
- [Evaluating Large Language Models at Evaluating Instruction Following](https://mlsanity.com/paper/2310.07641/) (2023)
- [ProcessBench: Identifying Process Errors in Mathematical Reasoning](https://mlsanity.com/paper/2412.06559/) (2024)
- [Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges](https://mlsanity.com/paper/2406.12624/) (2024)
- [Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems](https://mlsanity.com/paper/2505.00212/) (2025)
- [Autonomous Evaluation and Refinement of Digital Agents](https://mlsanity.com/paper/2404.06474/) (2024)
- [AgenTracer: Who Is Inducing Failure in the LLM Agentic Systems?](https://mlsanity.com/paper/2509.03312/) (2025)

## Built on by (most cited first, 2 of 2)

- [Self-Designed Evaluators and Warm Memory for Long-Horizon Agents](https://mlsanity.com/paper/2609.33717/) (2026)
- [Auditing Failure Attribution: Label Predictability and Judge Behavior](https://mlsanity.com/paper/2610.17747/) (2026)

## Abstract (arXiv)

A direct test of an LLM judge of agent trajectories injects faults into correct runs and reports recall, per fault type or by whether the fault broke the environment outcome (loud) or not (silent). Such recall can credit a judge with detection it does not have; paired discrimination, its flag rate on the faults minus its rate on the clean runs they came from, exposes this. Our testbed, a deterministic support desk with a scripted oracle and a one-step fault injector, labels all 400 trajectories exactly. A 14B judge shown only the request and final reply scores 34% to 76% recall on four fault types that leave the reply unchanged. There its input is the clean run's, so its paired discrimination is zero and that recall is its flag rate on clean runs. Splitting by outcome survival does not fix this: its loud recall of 84% is a paired +0.393 and its silent recall of 45% a paired +0.048, all from the two fault types that change the reply. Told to check each step, the same model flags every fault of those four types and 0 of 100 clean runs (95% CI up to 3.6%). It does not reliably check the reply: of four invented promises it flags one every time and the other three once in 42 faults. Shown every step but asked only about the reply, it still reaches a paired +0.69 on reply-unchanged faults, against +1.00 when told to check each step. We recommend reporting paired discrimination against clean parents, split by whether the fault reaches the judge's input and by outcome survival, and release the testbed, raw verdicts and analysis pipeline.