# Learning Probabilistic Responsibility Allocations for Multi-Agent Interactions

arXiv: https://arxiv.org/abs/2604.13128 · HTML: https://mlsanity.com/paper/2604.13128/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 32 references that are in the map.
- First posted on arXiv 13 Apr 2026.

## Builds on (most cited first, 24 of 24)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [nuScenes: A multimodal dataset for autonomous driving](https://mlsanity.com/paper/1903.11027/) (2019)
- [Categorical Reparameterization with Gumbel-Softmax](https://mlsanity.com/paper/1611.01144/) (2016)
- [Scalability in Perception for Autonomous Driving: Waymo Open Dataset](https://mlsanity.com/paper/1912.04838/) (2019)
- [Tutorial on Variational Autoencoders](https://mlsanity.com/paper/1606.05908/) (2016)
- [GPT-Driver: Learning to Drive with GPT](https://mlsanity.com/paper/2310.01415/) (2023)
- [AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting](https://mlsanity.com/paper/2103.14023/) (2021)
- [Control Barrier Function Based Quadratic Programs for Safety Critical Systems](https://mlsanity.com/paper/1609.06408/) (2016)
- [Human Motion Trajectory Prediction: A Survey](https://mlsanity.com/paper/1905.06113/) (2019)
- [Driving with LLMs: Fusing Object-Level Vector Modality for Explainable Autonomous Driving](https://mlsanity.com/paper/2310.01957/) (2023)
- [PRECOG: PREdiction Conditioned On Goals in Visual Multi-Agent Settings](https://mlsanity.com/paper/1905.01296/) (2019)
- [Equinox: neural networks in JAX via callable PyTrees and filtered transformations](https://mlsanity.com/paper/2111.00254/) (2021)
- [Embodied Understanding of Driving Scenarios](https://mlsanity.com/paper/2403.04593/) (2024)
- [Multimodal Probabilistic Model-Based Planning for Human-Robot Interaction](https://mlsanity.com/paper/1710.09483/) (2017)
- [Multimodal Deep Generative Models for Trajectory Prediction: A Conditional Variational Autoencoder Approach](https://mlsanity.com/paper/2008.03880/) (2020)
- [Social Coordination and Altruism in Autonomous Driving](https://mlsanity.com/paper/2107.00200/) (2021)
- [Cooperative Autonomous Vehicles that Sympathize with Human Drivers](https://mlsanity.com/paper/2107.00898/) (2021)
- [Trajectory Prediction Meets Large Language Models: A Survey](https://mlsanity.com/paper/2506.03408/) (2025)
- [Learning Responsibility Allocations for Safe Human-Robot Interaction with Applications to Autonomous Driving](https://mlsanity.com/paper/2303.03504/) (2023)
- [On the Differentiability of the Primal-Dual Interior-Point Method](https://mlsanity.com/paper/2406.11749/) (2024)
- [Responsibility-associated Multi-agent Collision Avoidance with Social Preferences](https://mlsanity.com/paper/2206.09030/) (2022)
- [Legible and Proactive Robot Planning for Prosocial Human-Robot Interactions](https://mlsanity.com/paper/2404.03734/) (2024)
- [Learning responsibility allocations for multi-agent interactions: A differentiable optimization approach with control barrier functions](https://mlsanity.com/paper/2410.07409/) (2024)
- [INTENT: Trajectory Prediction Framework with Intention-Guided Contrastive Clustering](https://mlsanity.com/paper/2503.04952/) (2025)

## Built on by (most cited first, 1 of 1)

- [Do as the Romans Do: Learning Universal Behaviors from Heterogeneous Agents](https://mlsanity.com/paper/2606.18537/) (2026)

## Abstract (arXiv)

Human behavior in interactive settings is shaped not only by individual objectives but also by shared constraints with others, such as safety. Understanding how people allocate responsibility, i.e., how much one deviates from their desired policy to accommodate others, can inform the design of socially compliant and trustworthy autonomous systems. In this work, we introduce a method for learning a probabilistic responsibility allocation model that captures the multimodal uncertainty inherent in multi-agent interactions. Specifically, our approach leverages the latent space of a conditional variational autoencoder, combined with techniques from multi-agent trajectory forecasting, to learn a distribution over responsibility allocations conditioned on scene and agent context. Although ground-truth responsibility labels are unavailable, the model remains tractable by incorporating a differentiable optimization layer that maps responsibility allocations to induced controls, which are available. We evaluate our method on the INTERACTION driving dataset and demonstrate that it not only achieves strong predictive performance but also provides interpretable insights, through the lens of responsibility, into patterns of multi-agent interaction.