SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks
Open accessAbstract
We introduce SwiftSage, a novel agent framework inspired by the dual-process theory of human cognition, designed to excel in action planning for complex interactive reasoning tasks. SwiftSage integrates the strengths of behavior cloning and prompting large language models (LLMs) to enhance task completion performance. The framework comprises two primary modules: the Swift module, representing fast and intuitive thinking, and the Sage module, emulating deliberate thought processes. The Swift module is a small encoder-decoder LM fine-tuned on the oracle agent's action trajectories, while the Sage module employs LLMs such as GPT-4 for subgoal planning and grounding. We develop a heuristic method to harmoniously integrate the two modules, resulting in a more efficient and robust problem-solving process. In 30 tasks from the ScienceWorld benchmark, SwiftSage significantly outperforms other methods such as SayCan, ReAct, and Reflexion, demonstrating its effectiveness in solving complex interactive tasks.
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
SWIFTSAGE is influenced by the dual-process theory of human cognition and its SWIFT module is a compact encoder-decoder language model fine-tuned using human-annotated datasets.
“SWIFTSAGE [82] is an agent influenced by the dual-process theory of human cognition [83], which is effective for solving complex interactive reasoning tasks.In this agent, the SWIFT module constitutes a compact encoder-decoder language model, which is fine-tuned using human-annotated datasets.”
From A survey on large language model based autonomous agents (2024) · cites this paper as [82]
The Abstract identifies SWIFTSAGE as inspired by dual-process theory and describes SWIFT as a small encoder-decoder language model. But B.1 Training Details of SWIFT says it was fine-tuned on 62,000 action-prediction examples, while the module description identifies searched oracle trajectories as the training data; these passages do not support the claim that the datasets were human-annotated. Page numbers are unavailable in our copy. (LLM, conf=0.90)
Experimental — accuracy not yet measuredSome agents evolve online not by updating their model weights, but by refining their plans and skill libraries on the fly.
“Some agents evolve online not by updating their model weights, but by refining their plans and skill libraries on the fly.”
From A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence (2025) · cites this paper as Lin et al., 2023
Section 3.3 says SAGE refines its plan on the fly when action history reveals a mistake or unachievable subgoal. The paper describes this as prompting an LLM during interaction, rather than updating weights, but it does not describe refining a skill library. It also reports that the separate SWIFT module was fine-tuned before use. (LLM, conf=0.87)
Experimental — accuracy not yet measuredSwiftSage operates with a fast-and-slow thinking process, reflecting on failures of its fast mode and switching to a deliberate, tool-using slow mode, adapting its strategy online based on task difficulty.
“Similarly, SwiftSage(Lin et al., 2023) operates with a fast-and-slow thinking process, where it can reflect on failures of its fast, intuitive mode and switch to a more deliberate, tool-using slow mode, adapting its strategy online based on task difficulty.”
From A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligence (2025) · cites this paper as Lin et al., 2023
Section 3.4 (page not provided) says SwiftSage switches from its fast SWIFT module to the deliberate SAGE module when it gets stuck, predicts an invalid action, faces a critical decision, or encounters an exception. The LLM section says SAGE can detect and fix mistakes. The copy does not describe SAGE as tool-using or say that switching is based on task difficulty; it describes a heuristic based on those specific triggers. (LLM, conf=0.85)
Experimental — accuracy not yet measured
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Cites
No references could be resolved yet.
Research works printed but not linked: 40
- Michael Ahn, Anthony Brohan, Noah Brown, Yevgen Chebotar, Omar Cortes, Byron David, Chelsea Finn, Keerthana Gopalakrishnan, Karol Hausman, Alexander Herzog, Daniel Ho, Jasmine Hsu, Julian Ibarz, Brian Ichter, Alex Irpan, Eric Jang, Rosario Jauregui Ruano, Kyle Jeffrey, Sally Jesmonth, Nikhil Jayant Joshi, Ryan C. Julian, Dmitry Kalashnikov, Yuheng Kuang, Kuang-Huei Lee, Sergey Levine, Yao Lu, Linda Luu, Carolina Parada, Peter Pastor, Jornell Quiambao, Kanishka Rao, Jarek Rettinghouse, Diego M Reyes, Pierre Sermanet, Nicolas Sievers, Clayton Tan, Alexander Toshev, Vincent Vanhoucke, F. Xia, Ted Xiao, Peng Xu, Sichun Xu, and Mengyuan Yan. Do as i can, not as i say: Grounding language in robotic affordances. In Conference on Robot Learning, 2022.
- Prithviraj Ammanabrolu and Matthew Hausknecht. Graph constrained reinforcement learning for natural language action spaces. In International Conference on Learning Representations, 2020.
- Prithviraj Ammanabrolu, Jack Urbanek, Margaret Li, Arthur D. Szlam, Tim Rocktaschel, and Jason Weston. How to motivate your dragon: Teaching goal-driven agents to speak and act in fantasy worlds. In North American Chapter of the Association for Computational Linguistics, 2020.
- Gabor Angeli, Melvin Johnson, and Christopher D. Manning. Leveraging linguistic structure for open domain information extraction. In Annual Meeting of the Association for Computational Linguistics, 2015.
- Thomas W. Anthony, Zheng Tian, and David Barber. Thinking fast and slow with deep learning and tree search. ArXiv, abs/1705.08439, 2017. · arXiv
- Tom B. Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, Sandhini Agarwal, Ariel Herbert-Voss, Gretchen Krueger, Tom Henighan, Rewon Child, Aditya Ramesh, Daniel M. Ziegler, Jeffrey Wu, Clemens Winter, Christopher Hesse, Mark Chen, Eric Sigler, Mateusz Litwin, Scott Gray, Benjamin Chess, Jack Clark, Christopher Berner, Sam McCandlish, Alec Radford, Ilya Sutskever, and Dario Amodei. Language models are few-shot learners. In Hugo Larochelle, Marc'Aurelio Ranzato, Raia Hadsell, Maria-Florina Balcan, and Hsuan-Tien Lin, editors, Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual, 2020.
- Sébastien Bubeck, Varun Chandrasekaran, Ronen Eldan, John A. Gehrke, Eric Horvitz, Ece Kamar, Peter Lee, Yin Tat Lee, Yuan-Fang Li, Scott M. Lundberg, Harsha Nori, Hamid Palangi, Marco Tulio Ribeiro, and Yi Zhang. Sparks of artificial general intelligence: Early experiments with gpt-4. ArXiv, abs/2303.12712, 2023. · arXiv
- Di Chen, Yiwei Bai, Wenting Zhao, Sebastian Ament, J. Gregoire, and Carla P. Gomes. Deep reasoning networks: Thinking fast and slow. ArXiv, abs/1906.00855, 2019. · arXiv
- Lili Chen, Kevin Lu, Aravind Rajeswaran, Kimin Lee, Aditya Grover, Michael Laskin, P. Abbeel, A. Srinivas, and Igor Mordatch. Decision transformer: Reinforcement learning via sequence modeling. In Neural Information Processing Systems, 2021.
- Hyung Won Chung, Le Hou, S. Longpre, Barret Zoph, Yi Tay, William Fedus, Eric Li, Xuezhi Wang, Mostafa Dehghani, Siddhartha Brahma, Albert Webson, Shixiang Shane Gu, Zhuyun Dai, Mirac Suzgun, Xinyun Chen, Aakanksha Chowdhery, Dasha Valter, Sharan Narang, Gaurav Mishra, Adams Wei Yu, Vincent Zhao, Yanping Huang, Andrew M. Dai, Hongkun Yu, Slav Petrov, Ed Huai hsin Chi, Jeff Dean, Jacob Devlin, Adam Roberts, Denny Zhou, Quoc V. Le, and Jason Wei. Scaling instruction-finetuned language models. ArXiv, abs/2210.11416, 2022. · arXiv
- Marc-Alexandre Côté, Ákos Kádár, Xingdi Yuan, Ben A. Kybartas, Tavian Barnes, Emery Fine, James Moore, Matthew J. Hausknecht, Layla El Asri, Mahmoud Adada, Wendy Tay, and Adam Trischler. Textworld: A learning environment for text-based games. In CGW@IJCAI, 2018.
- M. B. Ganapini, Murray Campbell, F. Fabiano, L. Horesh, Jonathan Lenchner, Andrea Loreggia, Nicholas Mattei, Francesca Rossi, Biplav Srivastava, and Kristen Brent Venable. Thinking fast and slow in ai: the role of metacognition. In International Conference on Machine Learning, Optimization, and Data Science, 2021.
- Yingqiang Ge, Wenyue Hua, Jianchao Ji, Juntao Tan, Shuyuan Xu, and Yongfeng Zhang. Openagi: When llm meets domain experts. arXiv, 2023.
- Ji He, Jianshu Chen, Xiaodong He, Jianfeng Gao, Lihong Li, Li Deng, and Mari Ostendorf. Deep reinforcement learning with a natural language action space. arXiv: Artificial Intelligence, 2015.
- Wenlong Huang, P. Abbeel, Deepak Pathak, and Igor Mordatch. Language models as zero-shot planners: Extracting actionable knowledge for embodied agents. ArXiv, abs/2201.07207, 2022. · arXiv
- Eric Kolve, Roozbeh Mottaghi, Winson Han, Eli VanderBilt, Luca Weihs, Alvaro Herrasti, Daniel Gordon, Yuke Zhu, Abhinav Gupta, and Ali Farhadi. AI2-THOR: An Interactive 3D Environment for Visual AI. arXiv, 2017.
- Bill Yuchen Lin, Chengsong Huang, Qianchu Liu, Wenda Gu, Sam Sommerer, and Xiang Ren. On grounded planning for embodied tasks with language models. ArXiv, abs/2209.00465, 2022. · arXiv
- Pan Lu, Baolin Peng, Hao Cheng, Michel Galley, Kai-Wei Chang, Ying Nian Wu, Song-Chun Zhu, and Jianfeng Gao. Chameleon: Plug-and-play compositional reasoning with large language models. ArXiv, abs/2304.09842, 2023. · arXiv
- Antoine Miech, Jean-Baptiste Alayrac, Ivan Laptev, Josef Sivic, and Andrew Zisserman. Think- ing fast and slow: Efficient text-to-visual retrieval with transformers. 2021 IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition (CVPR), pages 9821-9831, 2021.
- Keerthiram Murugesan, Mattia Atzeni, Pavan Kapanipathi, Pushkar Shukla, Sadhana Kumaravel, Gerald Tesauro, Kartik Talamadupula, Mrinmaya Sachan, and Murray Campbell. Text-based RL Agents with Commonsense Knowledge: New Challenges, Environments and Baselines. In Thirty Fifth AAAI Conference on Artificial Intelligence, 2021.
- Kolby Nottingham, Prithviraj Ammanabrolu, Alane Suhr, Yejin Choi, Hannaneh Hajishirzi, Sameer Singh, and Roy Fox. Do embodied agents dream of pixelated sheep: Embodied decision making using language guided world modelling. In International Conference on Machine Learning (ICML), 2023.
- Maxwell Nye, Michael Henry Tessler, Joshua B. Tenenbaum, and Brenden M. Lake. Improv- ing coherence and consistency in neural sequence models with dual-system, neuro-symbolic reasoning. In Neural Information Processing Systems, 2021.
- Long Ouyang, Jeff Wu, Xu Jiang, Diogo Almeida, Carroll L. Wainwright, Pamela Mishkin, Chong Zhang, Sandhini Agarwal, Katarina Slama, Alex Ray, John Schulman, Jacob Hilton, Fraser Kelton, Luke E. Miller, Maddie Simens, Amanda Askell, Peter Welinder, Paul Francis Christiano, Jan Leike, and Ryan J. Lowe. Training language models to follow instructions with human feedback. ArXiv, abs/2203.02155, 2022. · arXiv
- Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 21(140):1-67, 2020.
- Scott Reed, Konrad Zolna, Emilio Parisotto, Sergio Gomez Colmenarejo, Alexander Novikov, Gabriel Barth-Maron, Mai Gimenez, Yury Sulsky, Jackie Kay, Jost Tobias Springenberg, Tom Eccles, Jake Bruce, Ali Razavi, Ashley D. Edwards, Nicolas Manfred Otto Heess, Yutian Chen, Raia Hadsell, Oriol Vinyals, Mahyar Bordbar, and Nando de Freitas. A generalist agent. ArXiv, abs/2205.06175, 2022. · arXiv
- Nils Reimers and Iryna Gurevych. Sentence-BERT: Sentence embeddings using Siamese BERT-networks. In Proceedings of the 2019 Conference on Empirical Methods in Natural Lan- guage Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 3982-3992, Hong Kong, China, 2019. Association for Computational Linguistics.
- Timo Schick, Jane Dwivedi-Yu, Roberto Dessì, Roberta Raileanu, Maria Lomeli, Luke Zettle- moyer, Nicola Cancedda, and Thomas Scialom. Toolformer: Language models can teach themselves to use tools. ArXiv, abs/2302.04761, 2023. · arXiv
- Yongliang Shen, Kaitao Song, Xu Tan, Dong Sheng Li, Weiming Lu, and Yue Ting Zhuang. Hugginggpt: Solving ai tasks with chatgpt and its friends in huggingface. ArXiv, abs/2303.17580, 2023. · arXiv
- Noah Shinn, Beck Labash, and Ashwin Gopinath. Reflexion: an autonomous agent with dynamic memory and self-reflection. ArXiv, abs/2303.11366, 2023. · arXiv
- Mohit Shridhar, Xingdi Yuan, Marc-Alexandre Côté, Yonatan Bisk, Adam Trischler, and Matthew J. Hausknecht. Alfworld: Aligning text and embodied environments for interactive learning. ArXiv, abs/2010.03768, 2020. · arXiv
- Chan Hee Song, Jiaman Wu, Clay Washington, Brian M. Sadler, Wei-Lun Chao, and Yu Su. Llm-planner: Few-shot grounded planning for embodied agents with large language models. ArXiv, abs/2212.04088, 2022. · arXiv
- Ilya Sutskever, Oriol Vinyals, and Quoc V. Le. Sequence to sequence learning with neural networks. ArXiv, abs/1409.3215, 2014. · arXiv
- Faraz Torabi, Garrett Warnell, and Peter Stone. Behavioral cloning from observation. ArXiv, abs/1805.01954, 2018. · arXiv
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. Attention is all you need. In Isabelle Guyon, Ulrike von Luxburg, Samy Bengio, Hanna M. Wallach, Rob Fergus, S. V. N. Vishwanathan, and Roman Garnett, editors, Advances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA, pages 5998-6008, 2017.
- Ruoyao Wang, Peter Alexander Jansen, Marc-Alexandre Côté, and Prithviraj Ammanabrolu. Scienceworld: Is your agent smarter than a 5th grader? In Conference on Empirical Methods in Natural Language Processing, 2022.
- Zekun Wang, Ge Zhang, Kexin Yang, Ning Shi, Wangchunshu Zhou, Shaochun Hao, Guangzheng Xiong, Yizhi Li, Mong Yuan Sim, Xiuying Chen, Qingqing Zhu, Zhenzhu Yang, Adam Nik, Qi Liu, Chenghua Lin, Shi Wang, Ruibo Liu, Wenhu Chen, Ke Xu, Dayiheng Liu, Yike Guo, and Jie Fu. Interactive natural language processing. ArXiv, 2023.
- Zihao Wang, Shaofei Cai, Anji Liu, Xiaojian Ma, and Yitao Liang. Describe, explain, plan and select: Interactive planning with large language models enables open-world multi-task agents. ArXiv, abs/2302.01560, 2023. · arXiv
- Peter C Wason and J St BT Evans. Dual processes in reasoning? Cognition, 3(2):141-154, 1974.
- Shunyu Yao, Rohan Rao, Matthew J. Hausknecht, and Karthik Narasimhan. Keep calm and explore: Language models for action generation in text-based games. ArXiv, abs/2010.02903, 2020. · arXiv
- Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik Narasimhan, and Yuan Cao. React: Synergizing reasoning and acting in language models. ArXiv, abs/2210.03629, 2022. · arXiv
Cited by
Sorted by citations
A survey on large language model based autonomous agents
Wang, L., Ma, C., Feng, X., Zhang, Z., Yang, H., Zhang, J., Chen, Z., Tang, J., Chen, X., Lin, Y., et al. (2024) · Frontiers of Computer Science · doi:10.1007/s11704-024-40231-1 · Cited by 1,941 · see citing works
MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework
Hong, S., Zhuge, M., Chen, J., Zheng, X., Cheng, Y., Zhang, C., Wang, J., Wang, Z., Yau, S. K. S., Lin, Z., et al. (2023) · arXiv · doi:10.48550/arxiv.2308.00352 · Cited by 214 · see citing works
Expel: LLM agents are experiential learners
Zhao, A., Huang, D., Xu, Q., Lin, M., Liu, Y.-J., & Huang, G. (2024) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v38i17.29936 · Cited by 153 · see citing works
Zhang, Z., Yao, Y., Zhang, A., Tang, X., Ma, X., He, Z., Wang, Y., Gerstein, M., Wang, R., Liu, G., et al. (2025) · ACM Computing Surveys · doi:10.1145/3719341 · Cited by 51 · see citing works
Liu, J., Hao, W., Cheng, K., & Jin, D. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110319 · Cited by 5 · see citing works
Gao, H.-A., Geng, J., Hua, W., Hu, M., Juan, X., Liu, H., Liu, S., Qiu, J., Qi, X., Wu, Y., et al. (2025) · arXiv · doi:10.48550/arxiv.2507.21046 · Cited by 3 · see citing works
SmoothAgent: Efficient Long-Horizon LLM-Based Agent Serving with Lookahead Context Engineering
Pan, Z., Wang, Q., Hu, Z., Chen, C., Guan, Y., Zhou, Y., Swanson, S., & Ding, Y. (2026) · arXiv · doi:10.48550/arxiv.2607.00151
Can LLMs Correct Themselves? A Benchmark of Self-Correction in LLMs
Tie, G., Yuan, Z., Zhao, Z., Hu, C., Gu, T., Zhang, R., Zhang, S., Wu, J., Tu, X., Jin, M., et al. (2025) · arXiv · doi:10.48550/arxiv.2510.16062
Lyfe Agents: Generative agents for low-cost real-time social interactions
Kaiya, Z., Naim, M., Kondic, J., Cortes, M., Ge, J., Luo, S., Yang, G. R., & Ahn, A. (2023) · arXiv · doi:10.48550/arxiv.2310.02172 · Cited by 2 · see citing works
- Authors
- Lin, B. Y., Fu, Y., Yang, K., Brahman, F., Huang, S., Bhagavatula, C., Ammanabrolu, P., Choi, Y., & Ren, X.
- Venue
- Advances in Neural Information Processing Systems 36
- Published
- 2023
- Pages
- 23813-23825
- ISSN
- 1049-5258 (ISSN-L)
- OpenAlex
- W7133207012
- Cited by
- 50
- References
- 0
- Citations by year
- 2026: 14 · 2025: 14 · 2024: 16 · 2023: 6
- FWCI
- 8.88
- Type
- conference paper
- Language
- en
- Open access
- green
- arxiv.org
- Licence
- cc-by
- Locations
- neural information processing systems (publishedVersion)
- arXiv (Cornell University) (submittedVersion, cc-by)
- arXiv (Cornell University) (cc-by)
- Funders
- National Science Foundation (2048211, W911NF-19)
- Office of the Director of National Intelligence
- Allen Institute for Artificial Intelligence
- Defense Advanced Research Projects Agency (W911NF‐19‐2‐0271, N660011924033, N66001-19-2-4031, N66001-19)
- Advanced Research Projects Agency
- Intelligence Advanced Research Projects Activity
- Naval Information Warfare Center Pacific
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › AI-based Problem Solving and Planning
- Physical Sciences › Computer Science › Artificial Intelligence › Psychiatry, Mental Health, Neuroscience
- Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence in Games
- Keywords
- generative AI agents, fast and slow thinking, interactive tasks, large language models, multi-step reasoning
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
