AI Paper Atlas

Research map / Large language models / Agents, code and math

SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks

Bill Yuchen Lin, Yicheng Fu, Karina Yang, Faeze Brahman, Shiyu Huang, Chandra Bhagavatula, Prithviraj Ammanabrolu, Yejin Choi, Xiang Ren

NeurIPS 2023 · Large language models › Agents, code and math · approach: Long-horizon LLM agents · 29 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add the Sage (GPT-4 planning and grounding) module to Swift-only agent (ablation)SwiftSage · ScienceWorld (30 tasks, overall)average score: 49.22 -> 62.22Improved
add Sage module (GPT-3.5) to Swift-only on short tasks (ablation)SwiftSage · ScienceWorld short tasksaverage score: 78.68 -> 72.81Hurt
add Sage module (GPT-3.5) to Swift-only on medium tasks (ablation)SwiftSage · ScienceWorld medium tasksaverage score: 32.90 -> 55.34Improved
add Sage module (GPT-3.5) to Swift-only on long tasks (ablation)SwiftSage · ScienceWorld long tasksaverage score: 35.55 -> 57.99Improved
include answers to Q1-Q3 in the grounding stage (ablation)Sage module · ScienceWorld short tasksaverage score: about 2 pointsImproved
swap GPT-4 for GPT-3.5-turbo as the LLMSwiftSage · ScienceWorld (30 tasks, overall)average score: 84.68 -> 62.22Hurt
swap GPT-4 for GPT-3.5-turbo as the LLMReAct · ScienceWorld (30 tasks, overall)average score: 36.43 -> 19.76Hurt
fast-slow SwiftSage instead of per-step LLM prompting (ReAct)GPT-4 agents · ScienceWorld (30 tasks, overall)tokens per action: 1971.03 -> 757.07Improved
balanced T5-large Swift module with sliding-window history instead of TDT (11B)behavior cloning agent · ScienceWorld (30 tasks, overall)average score: 14.66 -> 49.22Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy