# Propose to Learn, Learn to Propose: Evaluability-Aware Assistance under Bounded Rationality

arXiv: https://arxiv.org/abs/2609.02242 · HTML: https://mlsanity.com/paper/2609.02242/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 23 of its 47 references that are in the map.
- First posted on arXiv 2 Sep 2026.

## Builds on (most cited first, 23 of 23)

- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Cooperative Inverse Reinforcement Learning](https://mlsanity.com/paper/1606.03137/) (2016)
- [Grounded Copilot: How Programmers Interact with Code-Generating Models](https://mlsanity.com/paper/2206.15000/) (2022)
- [Eliciting Human Preferences with Language Models](https://mlsanity.com/paper/2310.11589/) (2023)
- [A Large-Scale Survey on the Usability of AI Programming Assistants: Successes and Challenges](https://mlsanity.com/paper/2303.17125/) (2023)
- [Asking Easy Questions: A User-Friendly Approach to Active Reward Learning](https://mlsanity.com/paper/1910.04365/) (2019)
- [Online Bayesian Goal Inference for Boundedly-Rational Planning Agents](https://mlsanity.com/paper/2006.07532/) (2020)
- [Shared Autonomy via Deep Reinforcement Learning](https://mlsanity.com/paper/1802.01744/) (2018)
- [LLM-Based Test-Driven Interactive Code Generation: User Study and Empirical Evaluation](https://mlsanity.com/paper/2404.10100/) (2024)
- [Learning the Preferences of Ignorant, Inconsistent Agents](https://mlsanity.com/paper/1512.05832/) (2015)
- [When to Show a Suggestion? Integrating Human Feedback in AI-Assisted Programming](https://mlsanity.com/paper/2306.04930/) (2023)
- [AAAR-1.0: Assessing AI's Potential to Assist Research](https://mlsanity.com/paper/2410.22394/) (2024)
- [Information-Theoretic Bounded Rationality](https://mlsanity.com/paper/1512.06789/) (2015)
- [Modeling the Mistakes of Boundedly Rational Agents Within a Bayesian Theory of Mind](https://mlsanity.com/paper/2106.13249/) (2021)
- [AssistanceZero: Scalably Solving Assistance Games](https://mlsanity.com/paper/2504.07091/) (2025)
- [Bounded Rationality for LLMs: Satisficing Alignment at Inference-Time](https://mlsanity.com/paper/2505.23729/) (2025)
- [Human and Machine: How Software Engineers Perceive and Engage with AI-Assisted Code Reviews Compared to Their Peers](https://mlsanity.com/paper/2501.02092/) (2025)
- [Best-Response Bayesian Reinforcement Learning with Bayes-adaptive POMDPs for Centaurs](https://mlsanity.com/paper/2204.01160/) (2022)
- [Zero-Shot Assistance in Sequential Decision Problems](https://mlsanity.com/paper/2202.07364/) (2022)
- [Observation Interference in Partially Observable Assistance Games](https://mlsanity.com/paper/2412.17797/) (2024)
- [Modeling Boundedly Rational Agents with Latent Inference Budgets](https://mlsanity.com/paper/2312.04030/) (2023)
- [Cooperative Bayesian Optimization for Imperfect Agents](https://mlsanity.com/paper/2403.04442/) (2024)
- [More Than Irrational: Modeling Belief-Biased Agents](https://mlsanity.com/paper/2511.12359/) (2025)

## Abstract (arXiv)

AI assistants often collaborate by proposing candidate edits, plans, or designs that users evaluate before adoption. Existing assistance methods focus on proposal quality or user-goal inference, often assuming that the user can reliably evaluate any proposal, which can fail in practice because of bounded rationality. We study evaluability-aware proposal planning, where proposals serve both as task interventions and as probes for learning latent preferences and evaluation constraints, where the resulting belief updates then guide later proposals. We formalise this setting as ProSE, a hidden-parameter sequential assistance problem, and instantiate it with a KL-regularised bounded-rational binary response model in which acceptance trades off value gain against a distance-dependent evaluability penalty. Analysing the planning consequence of this likelihood reveals that likely accepted proposals and informative probes need not coincide, which explains why planners that only pursue acceptance systematically underperform. We operationalise ProSE with textscProSE-Plan, a depth-2 Bayes-adaptive planner that scores proposals by possible responses and response-induced posterior beliefs. In controlled graph simulations, textscProSE-Plan improves over evaluability-unaware and myopic baselines when evaluation cost is the bottleneck, and a probe-commit ablation confirms that our approach selects informative proposals that simpler methods miss. Our results thus identify user evaluability as a planning-relevant dimension of AI assistance, complementary to generation quality and preference inference.