GAIN: A Benchmark for Goal-Aligned Decision-Making of Large Language Models under Imperfect Norms

2026-03-19 · Masayuki Kawarada, Kodai Watanabe, Soichiro Murakami arxiv

We introduce GAIN (Goal-Aligned Decision-Making under Imperfect Norms), a benchmark designed to evaluate how large language models (LLMs) balance adherence to norms against business goals. Existing benchmarks typically focus on abstract scenarios rather than real-world business applications. Furthermore, they provide limited insights into the factors influencing LLM decision-making. This restricts their ability to measure models' adaptability to complex, real-world norm-goal conflicts. In GAIN, models receive a goal, a specific situation, a norm, and additional contextual pressures. These pressures, explicitly designed to encourage potential norm deviations, are a unique feature that differentiates GAIN from other benchmarks, enabling a systematic evaluation of the factors influencing decision-making. We define five types of pressures: Goal Alignment, Risk Aversion, Emotional/Ethical Appeal, Social/Authoritative Influence, and Personal Incentive. The benchmark comprises 1,200 scenarios across four domains: hiring, customer support, advertising and finance. Our experiments show that advanced LLMs frequently mirror human decision-making patterns. However, when Personal Incentive pressure is present, they diverge significantly, showing a strong tendency to adhere to norms rather than deviate from them.

📄 PDF Abstract BibTeX arXiv:2603.18469

Code (0)

등록된 구현이 없습니다.

Requirements for Aligned, Dynamic Resolution of Conflicts in Operational Constraints

2025-11-14 · Steven J. Jones, Robert E. Wray, John E. Laird arxiv

Deployed, autonomous AI systems must often evaluate multiple plausible courses of action (extended sequences of behavior) in novel or under-specified contexts. Despite extensive training, these systems will inevitably en…

Decision Making

Learning Transferable Latent User Preferences for Human-Aligned Decision Making

2026-05-12 · Alina Hyk, Sandhya Saisubramanian arxiv

Large language models (LLMs) are increasingly used as reasoning modules in many applications. While they are efficient in certain tasks, LLMs often struggle to produce human-aligned solutions. Human-aligned decision maki…

Decision Making

Hierarchical Decision Making with Structured Policies: A Principled Design via Inverse Optimization

2026-06-27 · Yuexuan Wang, Jingyuan Zhou, Kaidi Yang arxiv

Hierarchical decision-making frameworks are pivotal for addressing complex control tasks, enabling agents to decompose intricate problems into manageable subgoals. Despite their promise, existing hierarchical policies fa…

Reinforcement Learning Collision Avoidance Decision Making

Goal-driven Bayesian Optimal Experimental Design for Robust Decision-Making Under Model Uncertainty

2026-05-25 · Jinwoo Go, Xiaoning Qian, Byung-Jun Yoon arxiv

Bayesian optimal experimental design (BOED) selects experiments to maximize information gain about model parameters. However, in decision-critical settings, reducing parameter uncertainty does not necessarily improve dow…

Differential Privacy and Fairness in Decisions and Learning Tasks: A Survey

2022-02-16 · Ferdinando Fioretto, Cuong Tran, Pascal Van Hentenryck, Keyu Zhu

This paper surveys recent work in the intersection of differential privacy (DP) and fairness. It reviews the conditions under which privacy and fairness may have aligned or contrasting goals, analyzes how and why DP may …

Fairness Privacy Preserving