# What Did the Agent Actually Do? Evidence-Grounded Oversight for Long-Horizon Agents

arXiv: https://arxiv.org/abs/2610.06406 · HTML: https://mlsanity.com/paper/2610.06406/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 16 of its 36 references that are in the map.
- First posted on arXiv 5 Oct 2026.

## Builds on (most cited first, 16 of 16)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [PaperBench: Evaluating AI's Ability to Replicate AI Research](https://mlsanity.com/paper/2504.01848/) (2025)
- [SWE-chat: Coding Agent Interactions From Real Users in the Wild](https://mlsanity.com/paper/2604.20779/) (2026)
- [From Agent Traces to Trust: A Survey of Evidence Tracing and Execution Provenance in LLM Agents](https://mlsanity.com/paper/2606.04990/) (2026)
- [KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation](https://mlsanity.com/paper/2604.08455/) (2026)
- [Ask or Assume? Uncertainty-Aware Clarification-Seeking in Coding Agents](https://mlsanity.com/paper/2603.26233/) (2026)
- [BenchGuard: Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks](https://mlsanity.com/paper/2604.24955/) (2026)
- [AgentLens: Revealing The Lucky Pass Problem in SWE-Agent Evaluation](https://mlsanity.com/paper/2605.12925/) (2026)
- [Automated Benchmark Auditing for AI Agents and Large Language Models](https://mlsanity.com/paper/2605.26079/) (2026)
- [EvoArena: Tracking Memory Evolution for Robust LLM Agents in Dynamic Environments](https://mlsanity.com/paper/2606.13681/) (2026)
- [DeNovoSWE: Scaling Long-Horizon Environments for Generating Entire Repositories from Scratch](https://mlsanity.com/paper/2606.10728/) (2026)
- [Completion neq Collaboration: Scaling Collaborative Effort with Agents](https://mlsanity.com/paper/2510.25744/) (2025)
- [π-Bench: Evaluating Proactive Personal Assistant Agents in Long-Horizon Workflows](https://mlsanity.com/paper/2605.14678/) (2026)
- [AI Agents Push Humans Out of the Loop](https://mlsanity.com/paper/2608.23642/) (2026)
- [TIDE: Proactive Multi-Problem Discovery via Template-Guided Iteration](https://mlsanity.com/paper/2606.04743/) (2026)

## Abstract (arXiv)

As agents take on long-horizon tasks, users shift from making individual decisions to overseeing autonomous execution. Yet the volume of agent activity and the fragmentation of supporting evidence make it difficult to determine which decisions warrant user verification. We study monitors that identify consequential decisions and locate evidence to help users assess their implications. We introduce AgentMonBench, a software-engineering benchmark comprising three subsets that cover two complementary dimensions: alignment between requirements and behavior, and awareness of consequential autonomous decisions for verification. To support these judgments, we propose the Evidence-Grounded Behavior Graph (EBG), a training-free method that groups source-linked evidence into behaviors and organizes their relationships into a graph. EBG presents task-oriented views of this graph to help monitors interpret behavior in context. Experiments across eight models show that EBG improves decision identification and evidence localization in most settings compared with direct access to the original context. Further experiments show that EBG's evidence-localization gains persist across input scales and hyperparameter settings, while real-world applications illustrate its practical value for human oversight.