发表机构
Soongsil University（崇实大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文探究LLM隐藏状态中查询侧能力需求的可线性解码性，提出TACIT框架分解能力需求，实验发现隐藏状态可高精度解码但显式分类不可靠，揭示“结构化但沉默”现象。
AI 中文摘要
可靠的工具使用不仅仅需要触发机制或将查询与API描述匹配。在选择特定工具之前，智能体必须首先推断用户查询所隐含的能力需求。本文研究这些查询侧能力需求是否能在生成之前从LLM隐藏表示中被线性解码，以及这种隐藏状态的可访问性与显式语言分类相比如何。我们提出了TACIT框架，该框架将外部需求沿三个基本轴分解：来源、转换和世界效应，定义了八种结构上不同的能力类别。使用来自基准测试、合成示例和新领域场景的1,600个平衡训练查询，我们在四个开源权重LLM家族的生成前隐藏状态上训练线性探针。我们的实证结果表明，细粒度的能力结构在所有模型中都能以高精度被线性解码。然而，关键的是，我们揭示了表示到语言化的差距：当被要求以自然语言显式分类相同查询时，这些模型的可靠性显著降低。这种脱节表明，关于所需外部能力的信息在LLM隐藏表示中是线性可访问的，但未被可靠地表达，我们将这种现象定义为“结构化但沉默”。
英文摘要
Reliable tool use requires more than triggering a mechanism or matching a query to an API description. Before selecting a specific tool, an agent must first infer the capability requirements implied by the user query. In this paper, we investigate whether these query-side capability requirements are linearly decodable from LLM hidden representations prior to generation, and how this hidden-state accessibility compares with explicit verbal classification. We introduce TACIT, a framework that decomposes external requirements along three fundamental axes: Source, Transformation, and World Effect, defining eight structurally distinct capability classes. Using 1,600 balanced training queries from benchmarks, synthetic examples, and new domain scenarios, we train linear probes on pre-generation hidden states from four open-weight LLM families. Our empirical results demonstrate that fine-grained capability structures are linearly decodable with high accuracy across all models. Crucially, however, we expose a representation-to-verbalization gap: these same models are significantly less reliable when asked to explicitly classify the same queries in natural language. This disconnect indicates that information about required external capabilities is linearly accessible in LLM hidden representations but not reliably expressed, a phenomenon we define as "structured but silent."
CommentsAccepted to AACL-IJCNLP 2026 Findings
