SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917424819863552 |
|---|---|
| author | Feng, Xiachong Jiang, Yi Feng, Xiaocheng Yin, Deyi Qin, Libo Ye, Yangfan Huang, Lei Ma, Weitao Gu, Yuxuan Qin, Chonghan Qin, Bing Kong, Lingpeng |
| author_facet | Feng, Xiachong Jiang, Yi Feng, Xiaocheng Yin, Deyi Qin, Libo Ye, Yangfan Huang, Lei Ma, Weitao Gu, Yuxuan Qin, Chonghan Qin, Bing Kong, Lingpeng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_18982 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution Feng, Xiachong Jiang, Yi Feng, Xiaocheng Yin, Deyi Qin, Libo Ye, Yangfan Huang, Lei Ma, Weitao Gu, Yuxuan Qin, Chonghan Qin, Bing Kong, Lingpeng Artificial Intelligence 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. |
| title | SAVOIR: Learning Social Savoir-Faire via Shapley-based Reward Attribution |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2604.18982 |