mlsanityData to 14 Oct 2026

Effective Human-AI Teams via Learned Natural Language Rules and Onboarding

arXiv 2311.01007 ↗First posted 2 Nov 2023cs.LG cs.AI cs.HCcited by 20

Not on the map. Why?

Abstract (arXiv)

People are relying on AI agents to assist them with various tasks. The human must know when to rely on the agent, collaborate with the agent, or ignore its suggestions. In this work, we propose to learn rules, grounded in data regions and described in natural language, that illustrate how the human should collaborate with the AI. Our novel region discovery algorithm finds local regions in the data as neighborhoods in an embedding space where prior human behavior should be corrected. Each region is then described using a large language model in an iterative and contrastive procedure. We then teach these rules to the human via an onboarding stage. Through user studies on object detection and question-answering tasks, we show that our method can lead to more accurate human-AI teams. We also evaluate our region discovery and description algorithms separately.

Key facts. Cited by 20 papers in the map (data through 14 Oct 2026). Builds on 64 of its 96 references that are in the map. First posted on arXiv 2 Nov 2023.

Builds on

64 of 96 references.

References · 64

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2020Language Models are Few-Shot Learnerscited by 32,419
  3. 2014Microsoft COCO: Common Objects in Contextcited by 17,608
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  5. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  6. 2020Measuring Massive Multitask Language Understandingcited by 7,963
  7. 2016"Why Should I Trust You?": Explaining the Predictions of Any Classifiercited by 5,070
  8. 2022Scaling Instruction-Finetuned Language Modelscited by 3,024
  9. 2018Can a Suit of Armor Conduct Electricity? A New Dataset for Open Book Question Answeringcited by 2,164
  10. 2018BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learningcited by 1,288
  11. 2018Model Cards for Model Reportingcited by 1,080
  12. 2020TweetEval: Unified Benchmark and Comparative Evaluation for Tweet Classificationcited by 417
38 more
  1. 2020Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performancecited by 399
  2. 2020Effect of Confidence and Explanation on Accuracy and Trust Calibration in AI-Assisted Decision Makingcited by 281
  3. 2020Consistent Estimators for Learning to Defer to an Expertcited by 271
  4. 2017Predict Responsibly: Improving Fairness and Accuracy by Learning to Defercited by 233
  5. 2022Measuring Progress on Scalable Oversight for Large Language Modelscited by 225
  6. 2020Evaluating Explainable AI: Which Algorithmic Explanations Help Users Predict Model Behavior?cited by 190
  7. 2018On Human Predictions with Explanations and Predictions of Machine Learning Models: A Case Study on Deception Detectioncited by 159
  8. 2022Domino: Discovering Systematic Errors with Cross-Modal Embeddingscited by 156
  9. 2022GitHub Copilot AI pair programmer: Asset or Liability?cited by 152
  10. 2016End-to-end Learning of Driving Models from Large-scale Video Datasetscited by 148
  11. 2020Proxy Tasks and Subjective Measures Can Be Misleading in Evaluating Explainable AI Systemscited by 139
  12. 2022Explanations Can Reduce Overreliance on AI Systems During Decision-Makingcited by 111
  13. 2019The Algorithmic Automation Problem: Prediction, Triage, and Human Effortcited by 101
  14. 2020"Why is 'Chicago' deceptive?" Towards Building Model-Driven Tutorials for Humanscited by 86
  15. 2021Towards a Science of Human-AI Decision Making: A Survey of Empirical Studiescited by 86
  16. 2022Distilling Model Failures as Directions in Latent Spacecited by 82
  17. 2019An Evaluation of the Human-Interpretability of Explanationcited by 71
  18. 2023Who Should Predict? Exact Algorithms For Learning to Defer to Humanscited by 70
  19. 2022Human-AI Collaboration via Conditional Delegation: A Case Study of Content Moderationcited by 67
  20. 2023Who Should I Trust: AI or Myself? Leveraging Human and AI Correctness Likelihood to Promote Appropriate Trust in AI-Assisted Decision-Makingcited by 67
  21. 2020DynaSent: A Dynamic Benchmark for Sentiment Analysiscited by 63
  22. 2021The Spotlight: A General Method for Discovering Systematic Errors in Deep Learning Modelscited by 63
  23. 2022When does dough become a bagel? Analyzing the remaining mistakes on ImageNetcited by 61
  24. 2021Understanding the Effect of Out-of-distribution Examples and Interactive Explanations on Human-AI Decision Makingcited by 58
  25. 2022Describing Differences between Text Distributions with Natural Languagecited by 54
  26. 2018What can AI do for me: Evaluating Machine Learning Interpretations in Cooperative Playcited by 53
  27. 2018Do Explanations make VQA Models more Predictable to a Human?cited by 52
  28. 2023In Search of Verifiability: Explanations Rarely Enable Complementary Performance in AI-Advised Decision Makingcited by 50
  29. 2022Sample Efficient Learning of Predictors that Complement Humanscited by 46
  30. 2022Discovering Bugs in Vision Models using Off-the-shelf Image Generation and Captioningcited by 45
  31. 2022ImageNet-X: Understanding Model Mistakes with Factor of Variation Annotationscited by 44
  32. 2021Wordcraft: a Human-AI Collaborative Editor for Story Writingcited by 41
  33. 2022Improving Human-AI Partnerships in Child Welfare: Understanding Worker Practices, Challenges, and Desires for Algorithmic Decision Supportcited by 39
  34. 2022Adaptive Testing of Computer Vision Modelscited by 39
  35. 2023Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationalescited by 37
  36. 2023Improving Human-AI Collaboration With Descriptions of AI Behaviorcited by 36
  37. 2023Mass-Producing Failures of Multimodal Systems with Language Modelscited by 36
  38. 2021Selective Regression Under Fairness Criteriacited by 34

50 most cited of 64 All 64 (JSON)

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 10; 2025: 6; 2026: 4.

20 papers.

Sort:

By area:

  1. 2024When combinations of humans and AI are useful: A systematic review and meta-analysiscited by 106
  2. 2024Position: Towards Bidirectional Human-AI Alignmentcited by 18
  3. 2024Contrastive Explanations That Anticipate Human Misconceptions Can Improve Human Decision-Making Skillscited by 14
  4. 2024Towards Optimizing Human-Centric Objectives in AI-Assisted Decision-Making With Offline Reinforcement Learningcited by 13
  5. 2024Modulating Language Model Experiences through Frictionscited by 13
  6. 2024Challenges in Human-Agent Communicationcited by 12
  7. 2025AI, Help Me Thinkunicodex2014but for Myself: Assisting People in Complex Decision-Making by Providing Different Kinds of Cognitive Supportcited by 12
  8. 2024Utilizing Human Behavior Modeling to Manipulate Explanations in AI-Assisted Decision Making: The Good, the Bad, and the Scarycited by 10
  9. 2024On Evaluating Explanation Utility for Human-AI Decision Making in NLPcited by 9
  10. 2024Don't be Fooled: The Misinformation Effect of Explanations in Human-AI Collaborationcited by 8
  11. 2025Explaining and Improving Information Complementarities in Multi-Agent Decision-makingcited by 6
  12. 2025Explainable Mapper: Charting LLM Embedding Spaces Using Perturbation-Based Explanation and Verification Agentscited by 4
8 more
  1. 2024A Dynamic Model of Performative Human-ML Collaboration: Theory and Empirical Evidencecited by 2
  2. 2026ComplLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-makingcited by 2
  3. 2025Enabling Rapid Shared Human-AI Mental Model Alignment via the After-Action Reviewcited by 1
  4. 2025Teaching People LLM's Errors and Getting it Rightcited by 1
  5. 2026Human-LLM Collaborative Feature Engineering for Tabular Datacited by 1
  6. 2025DeCoDe: Defer-and-Complement Decision-Making via Decoupled Concept Bottleneck Modelscited by 0
  7. 2026TRIBE: Predicting Team Performance via Communication Behavior Ensemblescited by 0
  8. 2026Beyond "ChatGPT Can Make Mistakes": Designing Interventions to Support Metacognitive Monitoring in AI-Assisted Workcited by 0