Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions
by University of Edinburgh / Allen AI · free · Last verified 2026-03-17
Evaluates Auto-GPT-style autonomous agents on online decision-making tasks, introducing standardized benchmarks for assessing multi-step planning, tool use, and goal completion. The paper analyzes failure modes, provides ablations, and discusses the gap between open-ended autonomy and reliable task completion in long-horizon settings.
https://arxiv.org/abs/2306.02224 ↗B
B—Above Average
Adoption: B+Quality: B+Freshness: BCitations: B+Engagement: F
Specifications
- License
- Open Access
- Pricing
- free
- Capabilities
- autonomous-agents, benchmarking, long-horizon-planning, decision-making
- Integrations
- Use Cases
- agent-evaluation, research, autonomous-task-completion
- API Available
- No
- Tags
- agents, autonomous, gpt-4, decision-making, long-horizon
- Added
- 2026-03-17
- Completeness
- 100%
Index Score
62.4Adoption
72
Quality
78
Freshness
64
Citations
72
Engagement
0
Put AI to work for your business
Deploy this paper alongside autonomous AaaS agents that handle tasks end-to-end — no babysitting required.