# Effective Explanations Support Planning Under Uncertainty

arXiv: https://arxiv.org/abs/2605.08406 · HTML: https://mlsanity.com/paper/2605.08406/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 26 references that are in the map.
- First posted on arXiv 8 May 2026.

## Builds on (most cited first, 7 of 7)

- [Do As I Can, Not As I Say: Grounding Language in Robotic Affordances](https://mlsanity.com/paper/2204.01691/) (2022)
- [Speaker-Follower Models for Vision-and-Language Navigation](https://mlsanity.com/paper/1806.02724/) (2018)
- [Are Machine Rationales (Not) Useful to Humans? Measuring and Improving Human Utility of Free-Text Rationales](https://mlsanity.com/paper/2305.07095/) (2023)
- [Predictive, scalable and interpretable knowledge tracing on structured domains](https://mlsanity.com/paper/2403.13179/) (2024)
- [A Communication-First Account of Explanation](https://mlsanity.com/paper/2505.03732/) (2025)
- [Characterizing Language Use in a Collaborative Situated Game](https://mlsanity.com/paper/2512.03381/) (2025)
- [Language and Experience: A Computational Model of Social Learning in Complex Tasks](https://mlsanity.com/paper/2509.00074/) (2025)

## Abstract (arXiv)

Explaining how to get from A to B can be challenging. It requires mentally simulating what the listener will do based on what they are told. To capture this process, we propose a computational model that converts utterances into action plans: a large language model translates an explanation into program-like guidance (a policy prior and value map), and a planning agent executes it under partial observability. We score explanations by the efficiency and reliability of the resulting paths, penalizing replanning. Across four preregistered experiments, we collect a corpus of 1,200 explanations over 24 maps, elicit helpfulness judgments, measure baseline navigation, and test behavior with explanations of differing quality. Higher-scored explanations are judged more helpful and improve navigation: participants with explanations outperform those without, and high-scoring explanations help more than low-scoring ones. Together, these results show procedural explanation as utility-guided communication shaped by how language can be grounded into action under uncertainty.