Research map / Learning theory and algorithms / Prediction and decision losses
Adaptive Contracts for Cost-Effective AI Delegation
Eden Saig, Tamar Garbuz, Ariel D. Procaccia, Inbal Talgam-Cohen, Jamie Tucker-Foltz
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 changed | Where | Measured | Effect |
|---|---|---|---|
| use adaptive (selective inspection) contract instead of non-adaptive contracts | contract design · AlpacaEval 2.0 | principal utility: %14 higher than closest non-adaptive baseline | Improved |
| randomized inspection with committed probabilities (CoMI) | adaptive contracts · theory | existence of optimal contract | Hurt |
| restrict to no commitment or negative inspection payments | randomized inspection · theory | equilibrium existence | Improved |
| allow randomized inspection under UMI/CoNI | adaptive contracts · theory instances | principal utility | Improved |
| increase inspection cost per test | optimal policy · SWE-Bench | optimal inspection policy | Mixed |
| choose 3 initial tests and 17 refined tests | test-suite design · SWE-Bench | cost to principal | Improved |
| remove MLRP assumption | adaptive contract optimization · theory | computational tractability | Hurt |
Closest papers by meaning
- Incentivizing Quality Text Generation via Statistical Contracts (NeurIPS 2024 · similarity 0.85)
- Contracting with a Learning Agent (NeurIPS 2024 · similarity 0.83)
- Pricing and Competition for Generative AI (NeurIPS 2024 · similarity 0.83)
- Contract Design Under Approximate Best Responses (ICML 2025 · similarity 0.82)
- Paid with Models: Optimal Contract Design for Collaborative Machine Learning (AAAI 2025 · similarity 0.82)
- Cost-of-Pass: An Economic Framework for Evaluating Language Models (ICLR 2026 · similarity 0.82)
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.
