# To Ask or Not to Ask: Learning to Require Human Feedback

arXiv: https://arxiv.org/abs/2510.08314 · HTML: https://mlsanity.com/paper/2510.08314/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Reward modeling and RLHF, one of 10 directions in LLM alignment / RLHF / preference optimisation.
- Builds on 28 of its 46 references that are in the map.
- First posted on arXiv 9 Oct 2025.

## Placed in

- LLM alignment / RLHF / preference optimisation / [Reward modeling and RLHF](https://mlsanity.com/alignment/reward-modeling-and-rlhf/)

## Builds on (most cited first, 28 of 28)

- [Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification](https://mlsanity.com/paper/1502.01852/) (2015)
- [FiLM: Visual Reasoning with a General Conditioning Layer](https://mlsanity.com/paper/1709.07871/) (2017)
- [A Survey on Bias and Fairness in Machine Learning](https://mlsanity.com/paper/1908.09635/) (2019)
- [ChestX-ray8: Hospital-scale Chest X-ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases](https://mlsanity.com/paper/1705.02315/) (2017)
- [CheXNet: Radiologist-Level Pneumonia Detection on Chest X-Rays with Deep Learning](https://mlsanity.com/paper/1711.05225/) (2017)
- [Towards Out-Of-Distribution Generalization: A Survey](https://mlsanity.com/paper/2108.13624/) (2021)
- [Does the Whole Exceed its Parts? The Effect of AI Explanations on Complementary Team Performance](https://mlsanity.com/paper/2006.14779/) (2020)
- [Consistent Estimators for Learning to Defer to an Expert](https://mlsanity.com/paper/2006.01862/) (2020)
- [Predict Responsibly: Improving Fairness and Accuracy by Learning to Defer](https://mlsanity.com/paper/1711.06664/) (2017)
- [Cross-Entropy Loss Functions: Theoretical Analysis and Applications](https://mlsanity.com/paper/2304.07288/) (2023)
- [TorchXRayVision: A library of chest X-ray datasets and models](https://mlsanity.com/paper/2111.00595/) (2021)
- [Learning to Complement Humans](https://mlsanity.com/paper/2005.00582/) (2020)
- [When combinations of humans and AI are useful: A systematic review and meta-analysis](https://mlsanity.com/paper/2405.06087/) (2024)
- [Differentiable Learning Under Triage](https://mlsanity.com/paper/2103.08902/) (2021)
- [Who Should Predict? Exact Algorithms For Learning to Defer to Humans](https://mlsanity.com/paper/2301.06197/) (2023)
- [Calibrated Learning to Defer with One-vs-All Classifiers](https://mlsanity.com/paper/2202.03673/) (2022)
- [Optimal strategies for reject option classifiers](https://mlsanity.com/paper/2101.12523/) (2021)
- [Learning to Defer to Multiple Experts: Consistent Surrogate Losses, Confidence Calibration, and Conformal Ensembles](https://mlsanity.com/paper/2210.16955/) (2022)
- [Sample Efficient Learning of Predictors that Complement Humans](https://mlsanity.com/paper/2207.09584/) (2022)
- [Designing Decision Support Systems Using Counterfactual Prediction Sets](https://mlsanity.com/paper/2306.03928/) (2023)
- [In Defense of Softmax Parametrization for Calibrated and Consistent Learning to Defer](https://mlsanity.com/paper/2311.01106/) (2023)
- [Realizable H-Consistent and Bayes-Consistent Loss Functions for Learning to Defer](https://mlsanity.com/paper/2407.13732/) (2024)
- [Towards Human-AI Complementarity with Prediction Sets](https://mlsanity.com/paper/2405.17544/) (2024)
- [Enhanced H-Consistency Bounds](https://mlsanity.com/paper/2407.13722/) (2024)
- [Human-level Performance On Automatic Head Biometrics In Fetal Ultrasound Using Fully Convolutional Neural Networks](https://mlsanity.com/paper/1804.09102/) (2018)
- [Human Expertise in Algorithmic Prediction](https://mlsanity.com/paper/2402.00793/) (2024)
- [A Causal Framework for Evaluating Deferring Systems](https://mlsanity.com/paper/2405.18902/) (2024)
- [Auditing for Human Expertise](https://mlsanity.com/paper/2306.01646/) (2023)

## Built on by (most cited first, 6 of 6)

- [Oversight Has a Capacity: Calibrating Agent Guards to a Subjective, Fatiguing Human](https://mlsanity.com/paper/2606.08919/) (2026)
- [Consistent Learning-to-Defer with Expert-Conditional Advice](https://mlsanity.com/paper/2603.14324/) (2026)
- [Concise and Logically Consistent Conformal Sets for Neuro-Symbolic Concept-Based Models](https://mlsanity.com/paper/2605.18202/) (2026)
- [Should I stay or should I show? Learning to selectively disclose information](https://mlsanity.com/paper/2609.39818/) (2026)
- [Auto-Formalizing Neuro-Symbolic Predictors](https://mlsanity.com/paper/2610.01519/) (2026)
- [A Query Is Not a Commitment: Learning to Correct Expert Answers in Online Deferral](https://mlsanity.com/paper/2610.07084/) (2026)

## Abstract (arXiv)

Developing decision-support systems that complement human performance in classification tasks remains an open challenge. A popular approach, Learning to Defer (LtD), allows a Machine Learning (ML) model to pass difficult cases to a human expert. However, LtD treats humans and ML models as mutually exclusive decision-makers, restricting the expert contribution to mere predictions. To address this limitation, we propose Learning to Ask (LtA), a new framework that handles both when and how to incorporate expert input in an ML model. LtA is based on a two-part architecture: a standard ML model and an enriched model trained with additional expert human feedback, with a formally optimal strategy for selecting when to query the enriched model. We provide two practical implementations of LtA: a sequential approach, which trains the models in stages, and a joint approach, which optimises them simultaneously. For the latter, we design surrogate losses with realisable-consistency guarantees. Our experiments with synthetic and real expert data demonstrate that LtA provides a more flexible and powerful foundation for effective human-AI collaboration.