-
Hi @airsplay , I note that the angle feature size is 4 in both the paper and the default parameter setting but 128 in `run\agent.bash`.
Is this a special design? :)
-
I have a model that uses Bert embeddings. Before commit https://github.com/allenai/allennlp/commit/425f2af310e1000453ba94d4206166cbca1a971b, it worked fine. After pulling the latest AllenNLP code, I g…
-
Many *ACL volumes did not submit tutorial abstracts. These should be in the Anthology as part of the permanent record, and also as a means of gathering the presentations.
Progress status (possibly …
-
Hi,
I am confused about the relation F1 in the paper.
For me, a triplet is correct only if the relation and all the tokens of head and tail are correct,
eg: (B-PERSON √, I-PERSON √, Born_in √, B…
-
[Elsevier Ei indexing](https://suggestor.ei.engineeringvillage.com) is important to many academics in China. We need a process for doing this as automatically as possible for *ACL and EMNLP proceeding…
-
**[update]** Upstream bug: https://github.com/adobe-fonts/source-sans-pro/issues/87.
The website font, Source Sans Pro, is actually missing italic glyphs for Cyrillic.
As seen on [Google Fonts](…
-
**Is your feature request related to a problem? Please describe.**
Every registrable class has a classmethod from_params. We override it in cases where we want an alternative way to construct a obj…
-
Thank you for sharing your excellent code!
I have some questions on the calculation of Purity metrics shown in Table 5 in the EMNLP'18 paper.
I could not find the part of purity in this code. When I…
-
Currently, CSRankings uses ACM SIGs as the basis of its categories. However, Emery is left making many editorial decisions. Then only the top 3 conferences from each area are allowed. There are many i…
-
Here is an important work for NLI task, which can be added:
A Decomposable Attention Model for Natural Language Inference [EMNLP 2016, by Google]
[open code](https://github.com/HsiaoYetGun/Decomposa…