State-Grounded Conditioning：让 LLM 智能体依据实时状态给出正确方向
State-Grounded Conditioning: Wrapping User-Facing LLM Agents Where Direction Depends on Live State
做游戏内或实时状态类对话智能体的朋友看看，三个包装器把接地准确率从 60% 多拉到 96.7%，延迟还压在 1.5 秒。
论文提出 State-Grounded Conditioning（SGC），针对依赖实时用户状态（如游戏对局、会话历史）的 LLM 智能体，定义了“方向漂移”这一新失败类型：任务看似完成但方向与当前状态不符。SGC 通过 Perception、Grounding、Interaction 三个包装器把状态控制外置为规则内核。在一个 200 场会话、约 1000 轮对话的游戏内教练智能体基准上，三个包装器全部启用后，轮级接地准确率从 Prompting 的 61.1% 和 PE-Agent 的 69.8% 提升到 96.7%，会级从 20.0%/26.5% 提升到 83.5%，Perception 包装器将首 token 平均延迟保持在 1.5 秒（PE-Agent 为 6.1 秒）。
State-Grounded Conditioning: Wrapping User-Facing LLM Agents Where Direction Depends on Live State
We introduce State-Grounded Conditioning (SGC), a design principle for user-facing LLM agents that must condition on live user state (game state, session history, live inventory), and a distinct failure class we call direction drift: task-complete responses whose chosen direction misaligns with the current state. SGC externalises state-dependent control into rule kernels over structured inputs and three primary state slices, via Perception, Grounding, and Interaction wrappers with explicit conditioning dependencies. We evaluate SGC on a 200-session anonymised benchmark ($\approx$1,000 assistant model turns) from an in-game conversational coaching agent that guides players through consecutive competitive matches, reporting mean first-token latency and five human-annotated dialogue-quality metrics that jointly cover factual grounding and coach-like guidance progression. The Perception wrapper holds mean first-token latency at 1.5s (vs. 6.1s for PE-Agent inside a production tool-use harness); enabling all three wrappers lifts turn-level grounded accuracy from 61.1%/69.8% (Prompting / PE-Agent) to 96.7% and session-level grounded accuracy from 20.0%/26.5% to 83.5%; session-level grounding-failure incidents drop by $\approx$78% relative to the strongest baseline. A cumulative ablation shows complementary incremental gains as the wrappers are added. These results inform approximate state-slice orthogonality, without establishing independent per-wrapper effects.
