Yang-bug-star

Numbers of entities mismatched

1

Hi, I found the total number of DBpedia entities in "/data/entity2entityId.pkl" is 64362. However, in the run.py, the total number of entities is set to be 64368. (also in subkg.pkl),...

Since I'm a new learner, I wonder what the mask4key and mask4movie.npy files stand for. Could anyone tell me what these two files are used for?

The paper claims that the movies in TOTAL_NOVEL_MOVIES do not appear in training corporus, while I found many mentioned id in this list also appear in the training corporus, such...

In my view, res_movie_recall means the generated response contains exactly the same items as the ground truth, but function response_movie_recall_cal doesn't correctly judge whether the generated response contains exactly the...

代码实现中似乎并没有用到推荐模块得到的候选商品集合，因为selection_cross_decoder用的是db_encoding，这用的是context出现的实体编码，推荐模块并没有用到，不知道推荐模块对于回复生成在哪里起作用了，是不是应该将db_encoding改成推荐模块排名前50的实体表示才是正确的，感觉代码实现与论文提到的使用来自推荐模块的候选商品集合不符合

In all_response_movie_recall_cal，将 非movie token的label mask为0，但是selection__label也可能是为0的，且后续没有对此作额外判断会造成指标计算误差

At line 690 in model.py, masked_for_selection_token corresponds to the range [1:] which ignores the start token while latent corresponds to the range [0:end) which take the start into account. Their...