diff --git a/MEETING.md b/MEETING.md
index e0fed0a..9eca272 100644
--- a/MEETING.md
+++ b/MEETING.md
@@ -39,7 +39,7 @@ L = Λ · N · W_turn(L)        # agentic, T_human≈0
 
 ![](figs/f4a_apc_loss.png)
 
-LMetric 56.9%、load_only 54.1%、capped 31.6% APC，远低于 79.6% 上界。23pp 缺口直接来自跨 instance 路由丢的 intra-session hit。
+LMetric 56.9%、load_only 54.1% APC，远低于 79.6% 上界。23pp 缺口直接来自跨 instance 路由丢的 intra-session hit。
 
 ### 静态 PD-disagg：D 侧 KV 容量墙
 
diff --git a/PAPER_OUTLINE.md b/PAPER_OUTLINE.md
index 3e50547..ba90fda 100644
--- a/PAPER_OUTLINE.md
+++ b/PAPER_OUTLINE.md
@@ -145,7 +145,7 @@ Round-robin 和 load-aware routing（如 LMetric, OSDI'26）最大化 instance 
 
 **Figure 4: Three baselines, three failure modes** — 拆成三个子图，分别放在 §3.1/§3.2/§3.3：
 
-§3.1 — APC 实测 vs 理论上界 79.6% (lmetric 56.9%, load_only 54.1%, capped 31.6%, sticky 77.2%, unified 79.4%)：
+§3.1 — APC 实测 vs 理论上界 79.6% (lmetric 56.9%, load_only 54.1%, sticky 77.2%, unified 79.4%)：
 ![F4a APC loss](figs/f4a_apc_loss.png)
 
 §3.2 — D 侧 KV pool 占用 vs per-request KV footprint，4P+4D 和 6P+2D 在 agentic regime 都穿过 90% 内存墙：