# Where Knowledge and Authority Sit Changes What an Agent Benchmark Can Resolve

arXiv: https://arxiv.org/abs/2607.02975 · HTML: https://mlsanity.com/paper/2607.02975/

## Key facts

- No citing papers in the map yet (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 25 of its 36 references that are in the map.
- First posted on arXiv 3 Jul 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, 25 of 25)

- [WebArena: A Realistic Web Environment for Building Autonomous Agents](https://mlsanity.com/paper/2307.13854/) (2023)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (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)
- [LLMs Get Lost In Multi-Turn Conversation](https://mlsanity.com/paper/2505.06120/) (2025)
- [AppWorld: A Controllable World of Apps and People for Benchmarking Interactive Coding Agents](https://mlsanity.com/paper/2407.18901/) (2024)
- [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://mlsanity.com/paper/2310.11667/) (2023)
- [TheAgentCompany: Benchmarking LLM Agents on Consequential Real World Tasks](https://mlsanity.com/paper/2412.14161/) (2024)
- [ToolSandbox: A Stateful, Conversational, Interactive Evaluation Benchmark for LLM Tool Use Capabilities](https://mlsanity.com/paper/2408.04682/) (2024)
- [The Complexity of Decentralized Control of Markov Decision Processes](https://mlsanity.com/paper/1301.3836/) (2013)
- [AI Agents That Matter](https://mlsanity.com/paper/2407.01502/) (2024)
- [FANToM: A Benchmark for Stress-testing Machine Theory of Mind in Interactions](https://mlsanity.com/paper/2310.15421/) (2023)
- [Adding Error Bars to Evals: A Statistical Approach to Language Model Evaluations](https://mlsanity.com/paper/2411.00640/) (2024)
- [Generative agent-based modeling with actions grounded in physical, social, or digital space using Concordia](https://mlsanity.com/paper/2312.03664/) (2023)
- [CollabLLM: From Passive Responders to Active Collaborators](https://mlsanity.com/paper/2502.00640/) (2025)
- [Tell Me More! Towards Implicit User Intention Understanding of Language Model Driven Agents](https://mlsanity.com/paper/2402.09205/) (2024)
- [QuestBench: Can LLMs ask the right question to acquire information in reasoning tasks?](https://mlsanity.com/paper/2503.22674/) (2025)
- [Mind the Sim2Real Gap in User Simulation for Agentic Tasks](https://mlsanity.com/paper/2603.11245/) (2026)
- [Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systems](https://mlsanity.com/paper/2603.01045/) (2026)
- [Systematic Failures in Collective Reasoning under Distributed Information in Multi-Agent LLMs](https://mlsanity.com/paper/2505.11556/) (2025)
- [Autonomous Agents for Collaborative Task under Information Asymmetry](https://mlsanity.com/paper/2406.14928/) (2024)
- [TeamBench: Evaluating Agent Coordination under Enforced Role Separation](https://mlsanity.com/paper/2605.07073/) (2026)
- [Beyond the All-in-One Agent: Benchmarking Role-Specialized Multi-Agent Collaboration in Enterprise Workflows](https://mlsanity.com/paper/2605.08761/) (2026)
- [CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement](https://mlsanity.com/paper/2606.05793/) (2026)
- [CollabSim: A CSCW-Grounded Methodology for Investigating Collaborative Competence of LLM Agents through Controlled Multi-Agent Experiments](https://mlsanity.com/paper/2606.06399/) (2026)

## Abstract (arXiv)

Most agent benchmarks put facts, tools and permissions behind one interface. Real organizations spread them across people. Incognita asks what happens when the task and success criterion stay fixed but access does not. We transform eighteen customer-service tasks into three settings: direct access, one known intermediary, and six role-isolated participants whose capabilities must be discovered. Across 864 trials with four models, social access reduced success for every model; the pre-specified intervals excluded zero for two. The latest tested model, gpt-5.6-sol, achieved the highest social-access success at 0.65, a 0.11 decrease from centralized indirect access with an interval that included zero. Exploratory comparisons separated five of six model pairs under social access, while neither centralized setting separated any at this sample size. In post-hoc task-blocked tests, three pairwise interaction p-values remained significant after multiplicity adjustment. A reference-relative reader associates the wider gaps with failures to obtain needed information. Because the data cannot distinguish ineffective requests by the evaluated agent from inaccurate replies by simulated participants, the reader's labels describe where trajectories stopped, not why models differed.