fix(benchmark): 修复ReMe评估中的模型配置和空值处理问题

- 移除了retrieve_memory调用中不需要的llm_config_name参数
- 修复了长字符串打印的换行格式问题
- 添加了eval_result为空时的初始化处理
- 在accuracy评估中加入了eval_model_name参数传递

benchmark/halumem/eval_reme.py

@@ -577,6 +577,8 @@ class MemoryIntegrityEvaluator:
             )
 
             # Build result record
+            if eval_result is None:
+                eval_result = {}
             integrity_result = {
                 **memory_point,
                 "integrity_score": eval_result.get("score"),
@@ -619,9 +621,12 @@ class MemoryAccuracyEvaluator:
                 dialogue=formatted_dialogue,
                 golden_memories=memory_points,
                 candidate_memory=memory,
+                eval_model_name=self.eval_model_name,
             )
 
             # Build result record
+            if eval_result is None:
+                eval_result = {}
             accuracy_result = {
                 "memory_content": memory.get("content", memory.get("memory_content", str(memory))),
                 "memory_id": memory.get("memory_id", ""),

benchmark/longmemeval/eval_longmemeval_reme.py

@@ -785,7 +785,10 @@ class LongMemEvalEvaluator:
         print(f"Samples per type: {self.config.samples_per_type} (-1 = all)")
         print(f"Questions to process: {total_questions} | Top-K: {self.config.top_k}")
         print(f"Max Concurrency: {self.config.max_concurrency}")
-        print(f"Summary Model: {self.config.reme_model_name} | Retrieve Model: {self.config.retrieve_model_name} | Eval Model: {self.config.eval_model_name}")
+        print(
+            f"Summary Model: {self.config.reme_model_name} | Retrieve Model: {self.config.retrieve_model_name} "
+            f"| Eval Model: {self.config.eval_model_name}",
+        )
         print(f"Algo Version: {self.config.algo_version}")
         print("=" * 80 + "\n")