AI Paper Atlas

Research map / Large language models / AI and society

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

Hussein Mozannar, Jimin J Lee, Dennis Wei, Prasanna Sattigeri, Subhro Das, David Sontag

NeurIPS 2023 · Large language models › AI and society · approach: Social-science view of AI evaluation · 4 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
onboarding with IntegrAI regions vs Human+AI without onboardingHuman-AI team · BDD user studyparticipant accuracy (%): 77.2 -> 82.6Improved
onboarding with DOMINO-based baseline regions vs no onboardingHuman-AI team · BDD user studyparticipant accuracy (%): 77.2 -> 80.4No clear effect
add AI-integration recommendations to onboardingOnboard(ours) · BDD user studyparticipant accuracy (%): 82.6 -> 79.9No clear effect
add AI-integration recommendations to onboardingOnboard(ours) · BDD user studytime per example (s): 5.936 -> 7.622Hurt
onboarding vs no onboardingHuman+AI · MMLU user studyparticipant accuracy (%): 75.0 -> 74.4No clear effect
show only AI-integration recommendationsHuman+AI · MMLU user studyparticipant accuracy (%): 75.0 -> 69.8Hurt
onboarding vs no onboardingHuman+AI · MMLU user studytime per example (s): 23.623 -> 20.977Improved
IntegrAI regions vs DOMINOregion discovery · BDDtest error (%): 18.9 -> 17.8Improved
IntegrAI regions vs DOMINOregion discovery · MMLUtest error (%): 48.1 -> 45.3Improved
IntegrAI regions vs DOMINOregion discovery · DynaSenttest error (%): 20.0 -> 20.2No clear effect
include examples outside the region in description (S- not empty, m=5) (ablation)IntegrAI-Describe · MS-COCOMETEOR: 24.3 -> 26.1Improved
more refinement iterations m=0 to m=5 (ablation)IntegrAI-Describe · MS-COCOSPICE: 43.8 -> 45.2Improved

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