From 40bf66f34c090043b9c8c1b9c39e1867b234c1df Mon Sep 17 00:00:00 2001
From: Guo Cheng <224264187+GuoCheng24@users.noreply.github.com>
Date: Wed, 2 Sep 2026 08:49:45 +0800
Subject: [PATCH] Make the data-leakage trigger reachable

should_check_data_leakage compares the metric with == 1.0 / == 0.0, so the
check it guards (check_data_leakage, on by default) effectively never runs:
a leaked AUC lands on 0.99997 and a leaked RMSE on 1.2e-08. Compare against
configurable bounds instead, chosen conservatively so the exact-1.0 and
exact-0.0 cases stay inside the new bounds and ordinary scores do not fire.
---
 agents/triggers.py | 7 +++++--
 config/__init__.py | 3 +++
 config/config.yaml | 3 +++
 3 files changed, 11 insertions(+), 2 deletions(-)

diff --git a/agents/triggers.py b/agents/triggers.py
index 83609b23..0617408b 100644
--- a/agents/triggers.py
+++ b/agents/triggers.py
@@ -14,10 +14,13 @@ def should_check_data_leakage(agent, node: SearchNode) -> bool:
     metric_value = node.metric.value
     maximize = agent.metric_maximize
 
+    # Exact float equality almost never fires in practice: a leaked AUC lands
+    # on 0.99997, a leaked RMSE on 1.2e-08. Compare against a threshold instead,
+    # keeping the old exact-1.0 / exact-0.0 cases inside the new bounds.
     if maximize:
-        is_extreme = (metric_value == 1.0)
+        is_extreme = (metric_value >= agent.acfg.leakage_max_threshold)
     else:
-        is_extreme = (metric_value == 0.0)
+        is_extreme = (metric_value <= agent.acfg.leakage_min_threshold)
 
     if is_extreme:
         logger.info(
diff --git a/config/__init__.py b/config/__init__.py
index 66a24b24..28737424 100644
--- a/config/__init__.py
+++ b/config/__init__.py
@@ -102,6 +102,9 @@ class AgentConfig:
     use_evolution: bool = True
     use_fusion: bool = True
     use_aggregation: bool = True
+    # Bounds for the data-leakage trigger (see agents/triggers.py).
+    leakage_max_threshold: float = 0.999
+    leakage_min_threshold: float = 1e-6
 @dataclass
 class ExecConfig:
     timeout: int
diff --git a/config/config.yaml b/config/config.yaml
index 6ef3c1cc..37f6af13 100644
--- a/config/config.yaml
+++ b/config/config.yaml
@@ -56,6 +56,9 @@ agent:
 
   # --- Submission check ---
   check_data_leakage: True
+  # A metric at or beyond these bounds triggers the data-leakage check.
+  leakage_max_threshold: 0.999
+  leakage_min_threshold: 1e-6
 
   # --- Code gen mode ---
   use_diff_mode: True                 # Multi-Mode Codegen