mlsanityData to 7 Oct 2026

ReDAct: Uncertainty-Aware Deferral for LLM Agents

arXiv 2604.07036 ↗First posted 8 Apr 2026cs.CL cs.LG cs.MAcited by 5

On the map: LLM agents and tool use → Medical and scientific domain agents

Abstract (arXiv)

Recently, LLM-based agents have become increasingly popular across many applications, including complex sequential decision-making problems. However, they inherit the tendency of LLMs to hallucinate, leading to incorrect decisions. In sequential settings, even a single mistake can irreversibly degrade the trajectory, making hallucinations an even bigger problem. Although larger LLMs hallucinate less, they incur a significantly higher per-token cost. In this paper, we address this tradeoff by proposing ReDAct (Reason-Defer-Act). In ReDAct, an agent is equipped with two LLMs: a small, cheap model used by default, and a large, more reliable but expensive model. When the predictive uncertainty of the small model exceeds a calibrated threshold, the decision is deferred to the large model. We evaluate our approach in text-based embodied environments such as ALFWorld and MiniGrid and show that deferring only about 15% of decisions to the large model can match the quality of using it exclusively, while significantly reducing inference costs.

Key facts. Cited by 5 papers in the map (data through 7 Oct 2026). Placed in Medical and scientific domain agents, one of 16 directions in LLM agents and tool use. Builds on 39 of its 45 references that are in the map. First posted on arXiv 8 Apr 2026.

Builds on

39 of 45 references.

References · 39

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2023Judging LLM-as-a-Judge with MT-Bench and Chatbot Arenacited by 8,393
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,355
  4. 2022PaLM: Scaling Language Modeling with Pathwayscited by 5,304
  5. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,416
  6. 2023Toolformer: Language Models Can Teach Themselves to Use Toolscited by 4,072
  7. 2023Reflexion: Language Agents with Verbal Reinforcement Learningcited by 4,064
  8. 2021On the Opportunities and Risks of Foundation Modelscited by 3,961
  9. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,335
  10. 2022Do As I Can, Not As I Say: Grounding Language in Robotic Affordancescited by 2,302
  11. 2023Voyager: An Open-Ended Embodied Agent with Large Language Modelscited by 2,081
  12. 2022Language Models (Mostly) Know What They Knowcited by 1,783
27 more
  1. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,197
  2. 2020ALFWorld: Aligning Text and Embodied Environments for Interactive Learningcited by 1,143
  3. 2022Inner Monologue: Embodied Reasoning through Planning with Language Modelscited by 1,057
  4. 2016End to End Learning for Self-Driving Carscited by 1,000
  5. 2019Aleatoric and Epistemic Uncertainty in Machine Learning: An Introduction to Concepts and Methodscited by 903
  6. 2020A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challengescited by 813
  7. 2023Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generationcited by 795
  8. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 791
  9. 2023Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedbackcited by 788
  10. 2019ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Taskscited by 773
  11. 2021A Survey of Uncertainty in Deep Neural Networkscited by 675
  12. 2024RouteLLM: Learning to Route LLMs with Preference Datacited by 633
  13. 2019SelectiveNet: A Deep Neural Network with an Integrated Reject Optioncited by 333
  14. 2023Minigrid & Miniworld: Modular & Customizable Reinforcement Learning Environments for Goal-Oriented Taskscited by 320
  15. 2023Robots That Ask For Help: Uncertainty Alignment for Large Language Model Plannerscited by 298
  16. 2023Generating with Confidence: Uncertainty Quantification for Black-box Large Language Modelscited by 286
  17. 2020Consistent Estimators for Learning to Defer to an Expertcited by 262
  18. 2020Unsupervised Quality Estimation for Neural Machine Translationcited by 210
  19. 2023Shifting Attention to Relevance: Towards the Predictive Uncertainty Quantification of Free-Form Large Language Modelscited by 147
  20. 2023Large Language Model Cascades with Mixture of Thoughts Representations for Cost-efficient Reasoningcited by 140
  21. 2023LM-Polygraph: Uncertainty Estimation for Language Modelscited by 130
  22. 2024Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraphcited by 105
  23. 2022Calibrated Learning to Defer with One-vs-All Classifierscited by 68
  24. 2025Uncertainty Quantification for LLMs through Minimum Bayes Risk: Bridging Confidence and Consistencycited by 30
  25. 2024From Risk to Uncertainty: Generating Predictive Uncertainty Measures via Bayesian Estimationcited by 27
  26. 2023Dirichlet-based Uncertainty Quantification for Personalized Federated Learning with Improved Posterior Networkscited by 8
  27. 2025ReProbe: Efficient Test-Time Scaling of Multi-Step Reasoning by Probing Internal States of Large Language Modelscited by 8

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026Think Short, Defer Smart, Act, and Repeat: Calibrated Reasoning and Uncertainty-Aware Deferral for Edge LLM Agentscited by 1
  2. 2026TACIT-Switch: Cost-Aware Model Escalation for LLM Agents from Censored Supervisioncited by 1
  3. 2026Uncertainty Quantification for LLM Agents: A Taxonomy, an Evaluation Protocol, and an Empirical Studycited by 0
  4. 2026What Can Component-Replacement Evidence Establish? A Critical Scoping Review of Local Decisions in LLM Agentscited by 0
  5. 2026System Switch: When Should a Fast Decision Model Stop and Think?cited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025Mitigating LLM Hallucination via Behaviorally Calibrated Reinforcement Learning
  2. 2024Rethinking LLM Uncertainty: A Multi-Agent Approach to Estimating Black-Box Model Uncertainty
  3. 2025MIRAGE-Bench: LLM Agent is Hallucinating and Where to Find Them
  4. 2025LLM-based Agents Suffer from Hallucinations: A Survey of Taxonomy, Methods, and Directions
  5. 2026GSAR: Typed Grounding for Hallucination Detection and Recovery in Multi-Agent LLMs
  6. 2024Enhancing Multi-Agent Consensus through Third-Party LLM Integration: Analyzing Uncertainty and Mitigating Hallucinations in Large Language Models
  7. 2025Thinking, Faithful and Stable: Mitigating Hallucinations in LLMs
  8. 2026OptArgus: A Multi-Agent System to Detect Hallucinations in LLM-based Optimization Modeling