EntCollabBench shows that today's LLM agents still struggle with delegation, context transfer, parameter grounding, workflow closure, and decision commitment when tested in a simulated enterprise with 11 role-specialized agents.
arXiv preprint arXiv:2407.05291 (2024)
7 Pith papers cite this work, alongside 1 external citations. Polarity classification is still indexing.
7
Pith papers citing it
1
external citations · external index
citation-role summary
background 3
citation-polarity summary
roles
background 3polarities
background 3representative citing papers
Open 4B and 8B visual web agents achieve state-of-the-art results on browser benchmarks by predicting actions from screenshots and instructions, outperforming similar open models and some closed larger-model agents, with full release of data and code planned.
PhoneBuddy combines real-app and mock-app RL after shared SFT, raising real-phone task success from 36.67% to 45.33% and AndroidWorld from 60.3% to 83.2%.
ChainWorld builds 347 chains from atomic OSWorld tasks and benchmarks four agents under single-turn and multi-turn protocols, reporting a maximum 31% completion rate with distinct failure profiles.
VLAA-GUI adds mandatory visual verifiers, multi-tier loop breakers, and on-demand search to GUI agents, reaching 77.5% on OSWorld and 61.0% on WindowsAgentArena with some models exceeding human performance.
A benchmark with 886 real-world field work tasks across factories, warehouses, and retail, plus an evaluation method for multimodal AI agents.
citing papers explorer
-
Beyond the All-in-One Agent: Benchmarking Role-Specialized Multi-Agent Collaboration in Enterprise Workflows
EntCollabBench shows that today's LLM agents still struggle with delegation, context transfer, parameter grounding, workflow closure, and decision commitment when tested in a simulated enterprise with 11 role-specialized agents.
-
MolmoWeb: Open Visual Web Agent and Open Data for the Open Web
Open 4B and 8B visual web agents achieve state-of-the-art results on browser benchmarks by predicting actions from screenshots and instructions, outperforming similar open models and some closed larger-model agents, with full release of data and code planned.
-
PhoneBuddy: Training Open Models for Agentic Phone Use
PhoneBuddy combines real-app and mock-app RL after shared SFT, raising real-phone task success from 36.67% to 45.33% and AndroidWorld from 60.3% to 83.2%.
-
ChainWorld: Composing Long-Horizon Desktop Workloads from Atomic OSWorld Tasks
ChainWorld builds 347 chains from atomic OSWorld tasks and benchmarks four agents under single-turn and multi-turn protocols, reporting a maximum 31% completion rate with distinct failure profiles.
-
VLAA-GUI: Knowing When to Stop, Recover, and Search, A Modular Framework for GUI Automation
VLAA-GUI adds mandatory visual verifiers, multi-tier loop breakers, and on-demand search to GUI agents, reaching 77.5% on OSWorld and 61.0% on WindowsAgentArena with some models exceeding human performance.
-
FieldWorkArena: Agentic AI Benchmark for Real Field Work Tasks
A benchmark with 886 real-world field work tasks across factories, warehouses, and retail, plus an evaluation method for multimodal AI agents.
- Signal-Driven Observation for Long-Horizon Web Agents