Thinking Fast and Laterally: Multi-Agentic Approach for Reasoning About Uncertain Emerging Events

Abstract

This paper introduces lateral thinking to implement System-2 reasoning capabilities in AI systems, focusing on anticipatory and causal reasoning under uncertainty. We present a framework for systematic generation and modeling of lateral thinking queries and evaluation datasets. We introduce Streaming agentic Lateral Thinking (SALT), a multi-agent framework designed to process complex, low-specificity queries in streaming data environments. SALT implements lateral thinking-inspired System-2 reasoning through a dynamic communication structure between specialized agents. Preliminary quantitative and qualitative evaluations indicate SALT's potential to outperform single-agent systems in handling complex reasoning tasks.

Cite

Text

Dernbach et al. "Thinking Fast and Laterally: Multi-Agentic Approach for Reasoning About Uncertain Emerging Events." NeurIPS 2024 Workshops: Sys2-Reasoning, 2024.

Markdown

[Dernbach et al. "Thinking Fast and Laterally: Multi-Agentic Approach for Reasoning About Uncertain Emerging Events." NeurIPS 2024 Workshops: Sys2-Reasoning, 2024.](https://mlanthology.org/neuripsw/2024/dernbach2024neuripsw-thinking/)

BibTeX

@inproceedings{dernbach2024neuripsw-thinking,
  title     = {{Thinking Fast and Laterally: Multi-Agentic Approach for Reasoning About Uncertain Emerging Events}},
  author    = {Dernbach, Stefan and Michel, Alejandro and Agarwal, Khushbu and Brissette, Christopher and Gupta, Geetika and Choudhury, Sutanay},
  booktitle = {NeurIPS 2024 Workshops: Sys2-Reasoning},
  year      = {2024},
  url       = {https://mlanthology.org/neuripsw/2024/dernbach2024neuripsw-thinking/}
}