Merge pull request #9 from Felix-liu0989/fix/searchagent-multiturn-config

[cfg,tool] fix: add SearchAgent multi_turn config fields to MultiTurn…

Author: NLPJCL

verl/trainer/config/rollout/rollout.yaml

@@ -209,6 +209,29 @@ multi_turn:
   # Number of repeat rollouts for each sample
   num_repeat_rollouts: null
 
+  # SearchAgent-Zero: schedule max search turns, e.g. "0:4,50:8"
+  turn_limit_schedule: null
+
+  # SearchAgent-Zero: compress long search results before feeding back to the model
+  enable_tool_response_summary: False
+  summary_result_separator: "\\n"
+  summary_temperature: 0.6
+  summary_top_p: 0.95
+  summary_top_k: 20
+  summary_max_tokens: 1024
+  summary_use_external_model: False
+  summary_external_base_urls: ""
+  summary_external_model: ""
+  summary_external_api_key: ""
+  summary_external_timeout: 60.0
+  summary_external_enable_thinking: False
+
+  # SearchAgent-Zero: max queries inside one search tool call
+  max_queries_per_tool_call: null
+
+  # SearchAgent-Zero: duplicate search result detection threshold
+  duplicate_search_result_overlap_threshold: 0.6666666666666666
+
 # support logging rollout prob for debugging purpose
 # "Truncated importance sampling" requires rollout log probs, set to True when turning on Truncated importance sampling
 calculate_log_probs: False

verl/workers/config/rollout.py

@@ -86,6 +86,23 @@ class MultiTurnConfig(BaseConfig):
     format: str = "hermes"
     num_repeat_rollouts: Optional[int] = None
 
+    # SearchAgent-Zero multi-turn search agent options
+    turn_limit_schedule: Optional[str] = None
+    enable_tool_response_summary: bool = False
+    summary_temperature: float = 0.6
+    summary_top_p: float = 0.95
+    summary_top_k: int = 20
+    summary_max_tokens: int = 1024
+    summary_use_external_model: bool = False
+    summary_external_base_urls: str = ""
+    summary_external_model: str = ""
+    summary_external_api_key: str = ""
+    summary_external_timeout: float = 60.0
+    summary_external_enable_thinking: bool = False
+    max_queries_per_tool_call: Optional[int] = None
+    duplicate_search_result_overlap_threshold: float = 2 / 3
+    summary_result_separator: str = "\\n"
+
 
 @dataclass
 class CustomAsyncServerConfig(BaseConfig):