贝叶斯智能体蒸馏进 8B 语言模型：学提问也学信念状态
Learning What to Ask and What to Believe: Distilling a Variational Bayesian Agent into Language Models
把贝叶斯提问策略蒸馏进 8B 模型，猜词从 32 次降到 3.4 次，比 70B 老师还管用，思路很新。
研究把基于变分贝叶斯的后验推断与期望信息增益（EIG）提问策略的轻量智能体作为教师，蒸馏进一个 8B 语言模型。学生在三个 Multi-Turn Puzzles 任务上，Circuit Decoding 成功率从 0.00 提升到 0.93，电影推荐归一化排名从 0.50 降到 0.10，猜词尝试次数从 32 降到 3.4，在两个可比任务上超过已有报告的最强语言模型。消融显示 EIG 监督持续优于随机提问，而 70B 语言模型教师并不能在下游学生表现上超过贝叶斯教师。该技能无法迁移到未见过的任务族。
Learning What to Ask and What to Believe: Distilling a Variational Bayesian Agent into Language Models
Interactive tasks such as eliciting a user's preferences or diagnosing a hidden mechanism require an agent to track its uncertainty over a latent state, ask informative questions and act on the answers. Language models have no explicit mechanism for this, whereas Bayesian information seeking provides one through posterior inference and expected information gain (EIG), and we ask whether it can be distilled into an 8B language model. Our teacher, a lightweight Bayesian agent with a sequential variational belief and EIG query selection, matches or approaches exact and particle references on three task families. We generate synthetic interactions with the teacher on episodes disjoint from the benchmark and fine-tune \Student{} to reproduce its queries and its belief state as a structured scratchpad. On three Multi-Turn Puzzles tasks the student raises Circuit Decoding from 0.00 to 0.93 success, lowers the normalized rank of the recommended film from 0.50 to 0.10 and cuts word guessing from 32 to 3.4 attempts, exceeding the strongest reported language models on the two comparable tasks. Exposing the belief state closes much of the gap where the decision is an implicit computation rather than a read-out of the interaction; in matched teacher contrasts, EIG supervision consistently outperforms random querying, whereas greater inference fidelity does not uniformly improve the distilled student; a 70B language-model teacher does not match the Bayesian teacher in downstream student performance, and the learned skill does not transfer to an unseen task family. Variational Bayesian information seeking can thus be amortized into a language model by transferring the belief state that drives its questions and decisions.