-
**Abstract:**
> The dominant sequence transduction models are based on complex recurrent or convolutional neural networks that include an encoder and a decoder. The best performing models also conn…
-
1. A collection of 3 models
2. A form with a 'collection' type, where 'data' is the collection of models
3. A plain form in that collection type, with 2 fields
4. The 1st model will be used for the…
-
Hi, does DeepSpeed-MII support fairseq's translation model, such as transformer.wmt16.en-de or transformer.wmt19.en-de? as no task translation listed in the `Supported Models and Tasks` section.
-
I used marian-server for pretrained model. It works well for small models but is too slow for big-transformer models. Is there any solution to using multithread for the translation with these models?
…
-
**Describe the bug**
A question on the implementation of the NOCT cell temeprature model equation used with the CEC performance models was presented in https://www.researchgate.net/publication/361463…
-
## ASR
- [ ] ASR2K: Speech Recognition for Around 2000 Languages without Audio https://arxiv.org/abs/2209.02842
- [x] Whisper: Whisper is a general-purpose speech recognition model. https://github…
-
### Feature Name
MiniCPM-v2.5
### Feature Description
Research about MiniCPM-v2.5
### Research Findings
MiniCPM-v2.5 is a Chinese language model developed by the Beijing Academy of Artificial Int…
-
### Feature request
The main feature request involves a New Trainer Subclass, similar to Seq2SeqTrainer, but suitable for Decoder-Only LM.
### Motivation
`Seq2SeqTrainer` provides a great abstracti…
skpig updated
1 month ago
-
This issue isn't a problem to be fixed, rather it is a record to keep track of undergraduate student's work on TREC 2024. In a series of comments, contributors can write about their work (at a high le…
-
Task: @krmeierzalf or @Lachmuth-ZALF Please review translations from EN to DE of Geonode metadata fields
Description: When you click on the attached link and click on the "Edit this File" option, the…