From b9b54c5fcd8f2d2541da53d7fbaa4a89e59b7daa Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=E6=96=B9=E5=BA=94?= Date: Tue, 3 Mar 2026 12:28:33 +0800 Subject: [PATCH] =?UTF-8?q?fix(benchmark):=20=E4=BF=AE=E5=A4=8DReMe?= =?UTF-8?q?=E8=AF=84=E4=BC=B0=E4=B8=AD=E7=9A=84=E6=A8=A1=E5=9E=8B=E9=85=8D?= =?UTF-8?q?=E7=BD=AE=E5=92=8C=E7=A9=BA=E5=80=BC=E5=A4=84=E7=90=86=E9=97=AE?= =?UTF-8?q?=E9=A2=98?= MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit - 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 | 5 +++++ benchmark/longmemeval/eval_longmemeval_reme.py | 5 ++++- 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/benchmark/halumem/eval_reme.py b/benchmark/halumem/eval_reme.py index 14c0cb64..8cc27e53 100644 --- a/benchmark/halumem/eval_reme.py +++ b/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", ""), diff --git a/benchmark/longmemeval/eval_longmemeval_reme.py b/benchmark/longmemeval/eval_longmemeval_reme.py index 2edc204c..c529b434 100644 --- a/benchmark/longmemeval/eval_longmemeval_reme.py +++ b/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")