BehaviorBench：用真实行为轨迹测试模型预测用户决策能力
Salesforce 搞了个 BehaviorBench，用 2000 个真人的决策记录考模型会不会猜人心，有个人历史能从 53% 干到 75%，做智能体的可以看看。
Salesforce Research 发布 BehaviorBench 基准，测试模型能否从真实公开决策记录预测一个人的下一步选择，替代传统模拟用户。数据覆盖 2,000 名用户，包含 141K Belief 实例和 1.48M Trade 实例。实验显示加入个人历史后准确率从约 53% 提升到最高 75%，其中完整画像利于预测整体立场，近期历史利于预测下一步行动。基准已在 Hugging Face 以 Apache-2.0 协议开放。
BehaviorBench: Modeling Real-World User Decisions from Behavioral Traces: https://t.co/tKvAlxGFIc
Useful agents need to understand the people they serve. BehaviorBench tests whether models can predict a person's next decision from real public decision records in place of simulated users.
→ 2,000 users, 141K Belief + 1.48M Trade instances → Personal history lifts accuracy from ~53% to up to 75% → Profiles help predict overall stance; recent history helps predict next actions → Openly available on Hugging Face (Apache-2.0)
#FutureOfAI #AgenticAI #EnterpriseAI
