|
|
@@ -243,24 +243,22 @@ async def ground_single_case(
|
|
|
result = dict(case_item)
|
|
|
title = case_item.get("title", "")[:20] or "untitled"
|
|
|
|
|
|
- # 处理 workflow (strategy) - 整体处理，保持上下文
|
|
|
- workflow = case_item.get("workflow")
|
|
|
- if isinstance(workflow, dict) and "steps" in workflow:
|
|
|
- steps = workflow.get("steps", [])
|
|
|
-
|
|
|
- # 检查是否有任何 step 包含 apply_to_draft
|
|
|
+ # 处理 fragments - 整体处理，保持上下文。只要存在 fragments，就不再读取 capabilities。
|
|
|
+ fragments = case_item.get("fragments")
|
|
|
+ has_fragments = isinstance(fragments, list) and bool(fragments)
|
|
|
+ if has_fragments:
|
|
|
has_draft = any(
|
|
|
- isinstance(step, dict) and "apply_to_draft" in step
|
|
|
- for step in steps
|
|
|
+ isinstance(frag, dict) and "apply_to_draft" in frag
|
|
|
+ for frag in fragments
|
|
|
)
|
|
|
|
|
|
if has_draft:
|
|
|
- # 收集所有 step 的关键词（用于 API 搜索）
|
|
|
+ # 收集所有 fragment 的关键词（用于 API 搜索）
|
|
|
if use_api:
|
|
|
all_keywords = []
|
|
|
- for step in steps:
|
|
|
- if isinstance(step, dict) and "apply_to_draft" in step:
|
|
|
- apply_to_draft = step.get("apply_to_draft", {})
|
|
|
+ for frag in fragments:
|
|
|
+ if isinstance(frag, dict) and "apply_to_draft" in frag:
|
|
|
+ apply_to_draft = frag.get("apply_to_draft", {})
|
|
|
for key in ["实质", "形式"]:
|
|
|
for draft_text in apply_to_draft.get(key, []):
|
|
|
all_keywords.extend(extract_keywords_from_draft(draft_text))
|
|
|
@@ -268,20 +266,20 @@ async def ground_single_case(
|
|
|
|
|
|
if all_keywords:
|
|
|
categories = await search_categories_by_keywords(all_keywords, top_k=5)
|
|
|
- workflow_compact_tree = build_compact_tree(categories)
|
|
|
- workflow_ref_paths = list(dict.fromkeys(
|
|
|
+ frag_compact_tree = build_compact_tree(categories)
|
|
|
+ frag_ref_paths = list(dict.fromkeys(
|
|
|
c["path"] for c in categories if c.get("path")
|
|
|
))
|
|
|
else:
|
|
|
- workflow_compact_tree = compact_tree or "[]"
|
|
|
- workflow_ref_paths = []
|
|
|
+ frag_compact_tree = compact_tree or "[]"
|
|
|
+ frag_ref_paths = []
|
|
|
else:
|
|
|
- workflow_compact_tree = compact_tree or "[]"
|
|
|
- workflow_ref_paths = []
|
|
|
+ frag_compact_tree = compact_tree or "[]"
|
|
|
+ frag_ref_paths = []
|
|
|
|
|
|
- # 整个 workflow 传给 LLM（保持上下文）
|
|
|
- draft = {"strategy": workflow}
|
|
|
- prompt = render_grounding_prompt(template, "strategy", draft, workflow_compact_tree, workflow_ref_paths)
|
|
|
+ # 复用 capability grounding 的 prompt/schema，只把数据源从 workflow step 换成 fragment。
|
|
|
+ draft = {"capabilities": fragments}
|
|
|
+ prompt = render_grounding_prompt(template, "capability", draft, frag_compact_tree, frag_ref_paths)
|
|
|
messages = [{"role": "user", "content": prompt}]
|
|
|
|
|
|
grounded, cost = await call_llm_with_retry(
|
|
|
@@ -291,39 +289,28 @@ async def ground_single_case(
|
|
|
temperature=0.1,
|
|
|
max_tokens=4000,
|
|
|
max_retries=3,
|
|
|
- schema_name="apply_to_grounding_strategy",
|
|
|
- task_name=f"Ground_W_{title}",
|
|
|
+ schema_name="apply_to_grounding_capability",
|
|
|
+ task_name=f"Ground_F_{title}",
|
|
|
)
|
|
|
total_cost += cost
|
|
|
|
|
|
- # 按 order 回填 apply_to
|
|
|
- if grounded and isinstance(grounded.get("strategy"), dict):
|
|
|
- grounded_steps = grounded["strategy"].get("steps", [])
|
|
|
- # 建立 order -> apply_to 的映射
|
|
|
- order_to_apply_to = {}
|
|
|
- for grounded_step in grounded_steps:
|
|
|
- if isinstance(grounded_step, dict):
|
|
|
- order = grounded_step.get("order")
|
|
|
- apply_to = grounded_step.get("apply_to")
|
|
|
- if order is not None and apply_to is not None:
|
|
|
- order_to_apply_to[order] = apply_to
|
|
|
-
|
|
|
- # 回填到原 steps
|
|
|
- updated_steps = []
|
|
|
- for step in steps:
|
|
|
- updated_step = dict(step)
|
|
|
- order = step.get("order")
|
|
|
- if order in order_to_apply_to:
|
|
|
- updated_step["apply_to"] = order_to_apply_to[order]
|
|
|
- updated_step.pop("apply_to_draft", None)
|
|
|
- updated_steps.append(updated_step)
|
|
|
-
|
|
|
- result["workflow"] = dict(workflow)
|
|
|
- result["workflow"]["steps"] = updated_steps
|
|
|
-
|
|
|
- # 处理 capabilities - 整体处理，保持上下文
|
|
|
+ # 按索引回填 apply_to。输入数组来自 fragments，输出数组使用 capability schema。
|
|
|
+ if grounded and isinstance(grounded.get("capabilities"), list):
|
|
|
+ grounded_frags = grounded["capabilities"]
|
|
|
+ updated_fragments = []
|
|
|
+ for idx, frag in enumerate(fragments):
|
|
|
+ updated_frag = dict(frag)
|
|
|
+ if idx < len(grounded_frags) and isinstance(grounded_frags[idx], dict):
|
|
|
+ apply_to = grounded_frags[idx].get("apply_to")
|
|
|
+ if apply_to is not None:
|
|
|
+ updated_frag["apply_to"] = apply_to
|
|
|
+ updated_frag.pop("apply_to_draft", None)
|
|
|
+ updated_fragments.append(updated_frag)
|
|
|
+ result["fragments"] = updated_fragments
|
|
|
+
|
|
|
+ # 没有 fragments 时，才回退处理 capabilities。
|
|
|
capabilities = case_item.get("capabilities")
|
|
|
- if isinstance(capabilities, list) and capabilities:
|
|
|
+ if not has_fragments and isinstance(capabilities, list) and capabilities:
|
|
|
has_draft = any(
|
|
|
isinstance(cap, dict) and "apply_to_draft" in cap
|
|
|
for cap in capabilities
|
|
|
@@ -430,20 +417,16 @@ async def apply_grounding(
|
|
|
# 过滤出需要处理的 case（有 apply_to_draft 的）
|
|
|
needs_grounding = []
|
|
|
for case in cases:
|
|
|
- workflow = case.get("workflow")
|
|
|
+ fragments = case.get("fragments")
|
|
|
capabilities = case.get("capabilities")
|
|
|
- # 检查 step 级别的 apply_to_draft
|
|
|
- has_workflow_draft = (
|
|
|
- isinstance(workflow, dict) and
|
|
|
- any(
|
|
|
- isinstance(step, dict) and "apply_to_draft" in step
|
|
|
- for step in workflow.get("steps", [])
|
|
|
- )
|
|
|
+ has_fragments = isinstance(fragments, list) and bool(fragments)
|
|
|
+ has_frag_draft = isinstance(fragments, list) and any(
|
|
|
+ isinstance(frag, dict) and "apply_to_draft" in frag for frag in fragments
|
|
|
)
|
|
|
- has_cap_draft = isinstance(capabilities, list) and any(
|
|
|
+ has_cap_draft = not has_fragments and isinstance(capabilities, list) and any(
|
|
|
isinstance(c, dict) and "apply_to_draft" in c for c in capabilities
|
|
|
)
|
|
|
- if has_workflow_draft or has_cap_draft:
|
|
|
+ if has_frag_draft or has_cap_draft:
|
|
|
needs_grounding.append(case)
|
|
|
|
|
|
print(f"Grounding apply_to for {len(needs_grounding)}/{len(cases)} cases...")
|
elksmmx