Skin-in-the-Game: Decision Making via Multi-Stakeholder Alignment in LLMs
Bilgehan Sel, Priya Shanmugasundaram, Mohammad Kachuee, Kun Zhou, Ruoxi Jia, Ming Jin
Abstract
Large Language Models (LLMs) have shown remarkable capabilities in tasks such as summarization, arithmetic reasoning, and question answering. However, they encounter significant challenges in the domain of moral reasoning and ethical decision-making, especially in complex scenarios with multiple stakeholders. This paper introduces the Skin-in-the-Game (SKIG) framework, aimed at enhancing moral reasoning in LLMs by exploring decisions' consequences from multiple stakeholder perspectives. Central to SKIG's mechanism is simulating accountability for actions, which, alongside empathy exercises and risk assessment, is pivotal to its effectiveness. We validate SKIG's performance across various moral reasoning benchmarks with proprietary and opensource LLMs, and investigate its crucial components through extensive ablation analyses. The code and related content can be found in: skin-inthe-game.github.io
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
6 top-tier papers have cited it since. Know what they found before your reviewers do.
Lune can draw on
the full paper · the top-tier papers that cite it · related work
Free to start. No credit card required.
Lineage
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- LLMs Can Plan Only If We Tell Them cites Tree of Thoughts: Deliberate Problem Solving with Large Language Models
- LLMs Can Plan Only If We Tell Them cites Large Language Models are Zero-Shot Reasoners
- LLMs Can Plan Only If We Tell Them cites Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
- LLMs Can Plan Only If We Tell Them cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Reinforcement Learning with Backtracking Feedback cites Tree of Thoughts: Deliberate Problem Solving with Large Language Models
- Reinforcement Learning with Backtracking Feedback cites LLMs Can Plan Only If We Tell Them
- Reinforcement Learning with Backtracking Feedback cites Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
- Reinforcement Learning with Backtracking Feedback cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- LLMs Can Reason Faster Only If We Let Them cites Tree of Thoughts: Deliberate Problem Solving with Large Language Models
- LLMs Can Reason Faster Only If We Let Them cites LLMs Can Plan Only If We Tell Them
- LLMs Can Reason Faster Only If We Let Them cites Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models
- LLMs Can Reason Faster Only If We Let Them cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Who Does What? Archetypes of Roles Assigned to LLMs During Human-AI Decision-Making cites Large Language Models are Zero-Shot Reasoners
- Who Does What? Archetypes of Roles Assigned to LLMs During Human-AI Decision-Making cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Large Language Models are Zero-Shot Reasoners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models cites Tree of Thoughts: Deliberate Problem Solving with Large Language Models
- Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language Models cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
- Finetuned Language Models are Zero-Shot Learners cites Chain-of-Thought Prompting Elicits Reasoning in Large Language Models
Cited by top-tier papers6
- Algorithm of Thoughts: Enhancing Exploration of Ideas in Large Language ModelsBilgehan Sel, Ahmad Al-Tawaha, Vanshaj Khattar, Ruoxi Jia et al.ICML 2024 · 108 citations
- Who Does What? Archetypes of Roles Assigned to LLMs During Human-AI Decision-MakingShreya Chappidi, Jatinder Singh, Andra Valentina KrauzeCHI 2026 · 2 citations
- LLMs Can Plan Only If We Tell ThemBilgehan Sel, Ruoxi Jia, Ming JinICLR 2025
- Reinforcement Learning with Backtracking FeedbackBilgehan Sel, Vaishakh Keshava, Phillip Wallis, Lukas Rutishauser et al.NeurIPS 2025
- LLMs Can Reason Faster Only If We Let ThemBilgehan Sel, Lifu Huang, Naren Ramakrishnan, Ruoxi Jia et al.ICML 2025
Builds on12
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Large Language Models are Zero-Shot ReasonersTakeshi Kojima, Shixiang Shane Gu, Machel Reid, Yutaka Matsuo et al.NeurIPS 2022 · 8,168 citations
- Measuring Massive Multitask Language UnderstandingDan Hendrycks, Collin Burns, Steven Basart, Andy Zou et al.ICLR 2021 · 7,905 citations
- Tree of Thoughts: Deliberate Problem Solving with Large Language ModelsShunyu Yao, Dian Yu, Jeffrey Zhao, Izhak Shafran et al.NeurIPS 2023 · 5,068 citations
- Finetuned Language Models are Zero-Shot LearnersJason Wei, Maarten Bosma, Vincent Y. Zhao, Kelvin Guu et al.ICLR 2022 · 4,966 citations
Related papers
- Do Morals Guide How LLMs Think? The Role of Ethical Perspectives in General Problem SolvingIseo Kim, Eunjin Hong, Juae KimACL 2026
- Structured Moral Reasoning in Language Models: A Value-Grounded Evaluation FrameworkMohna Chakraborty, Lu Wang, David JurgensEMNLP 2025
- SOLAR: Towards Characterizing Subjectivity of Individuals through Modeling Value Conflicts and Trade-offsYounghun Lee, Dan GoldwasserEMNLP 2025
- Unveiling the Bias Impact on Symmetric Moral Consistency of Large Language ModelsZiyi Zhou, Xinwei Guo, Jiashi Gao, Xiangyu Zhao et al.NeurIPS 2024
- Knowing-but-Doing: Diagnosing and Defending Role-Play-Driven LLMs Jailbreaks via Moral DisengagementHaiming Qin, Jianxun Lian, Qimin Zhong, Mingyang Zhou et al.ACL Findings 2026 · 1 citation
