# SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution

arXiv: https://arxiv.org/abs/2604.18982 · HTML: https://mlsanity.com/paper/2604.18982/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 25 references that are in the map.
- First posted on arXiv 21 Apr 2026.

## Builds on (most cited first, 15 of 15)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [SOTOPIA: Interactive Evaluation for Social Intelligence in Language Agents](https://mlsanity.com/paper/2310.11667/) (2023)
- [Understanding Social Reasoning in Language Models with Language Models](https://mlsanity.com/paper/2306.15448/) (2023)
- [Plug-and-Play Policy Planner for Large Language Model Powered Dialogue Agents](https://mlsanity.com/paper/2311.00262/) (2023)
- [Emergent Social Learning via Multi-agent Reinforcement Learning](https://mlsanity.com/paper/2010.00581/) (2020)
- [SDPO: Segment-Level Direct Preference Optimization for Social Agents](https://mlsanity.com/paper/2501.01821/) (2025)
- [Sotopia-RL: Reward Design for Social Intelligence](https://mlsanity.com/paper/2508.03905/) (2025)
- [A Survey on Large Language Model-Based Social Agents in Game-Theoretic Scenarios](https://mlsanity.com/paper/2412.03920/) (2024)
- [Towards Social AI: A Survey on Understanding Social Interactions](https://mlsanity.com/paper/2409.15316/) (2024)
- [SOTOPIA-Omega: Dynamic Strategy Injection Learning and Social Instruction Following Evaluation for Social Agents](https://mlsanity.com/paper/2502.15538/) (2025)
- [Reasoning Does Not Necessarily Improve Role-Playing Ability](https://mlsanity.com/paper/2502.16940/) (2025)
- [Dialogue Action Tokens: Steering Language Models in Goal-Directed Dialogue with a Multi-Turn Planner](https://mlsanity.com/paper/2406.11978/) (2024)
- [Social Intelligence Data Infrastructure: Structuring the Present and Navigating the Future](https://mlsanity.com/paper/2403.14659/) (2024)

## Built on by (most cited first, 4 of 4)

- [SocialRL: Refining LLMs' Social Intelligence through Multi-turn Reinforcement Learning and Reward Design](https://mlsanity.com/paper/2609.09764/) (2026)
- [Element-Aware Group Learning for E-Commerce Image Generation](https://mlsanity.com/paper/2608.00584/) (2026)
- [RewardWeaver: Long-Horizon Interactive Learning for Language Agents via Self-Evolving Reward Adaptation](https://mlsanity.com/paper/2610.10120/) (2026)
- [What Makes a Social Agent's Interactions Effective? Learning from Verifiable Outcomes in Paired Worlds](https://mlsanity.com/paper/2610.20195/) (2026)

## Abstract (arXiv)

Social intelligence, the ability to navigate complex interpersonal interactions, presents a fundamental challenge for language agents. Training such agents via reinforcement learning requires solving the credit assignment problem: determining how individual utterances contribute to multi-turn dialogue outcomes. Existing approaches directly employ language models to distribute episode-level rewards, yielding attributions that are retrospective and lack theoretical grounding. We propose SAVOIR (ShApley Value fOr SocIal RL), a novel principled framework grounded in cooperative game theory. Our approach combines two complementary principles: expected utility shifts evaluation from retrospective attribution to prospective valuation, capturing an utterance's strategic potential for enabling favorable future trajectories; Shapley values ensure fair credit distribution with axiomatic guarantees of efficiency, symmetry, and marginality. Experiments on the SOTOPIA benchmark demonstrate that SAVOIR achieves new state-of-the-art performance across all evaluation settings, with our 7B model matching or exceeding proprietary models including GPT-4o and Claude-3.5-Sonnet. Notably, even large reasoning models consistently underperform, suggesting social intelligence requires qualitatively different capabilities than analytical reasoning.