-
This is a possible solution to #418
Our model aims to predict the optimal `split_every` value that makes the reduction as fast as possible.
This parameter affects the input data size of each func…
-
I fill "stt_en_fastconformer_ctc_small" in init_from_pretrained_model but i get an error "not found" the "stt_en_fastconformer_ctc_small"
```yaml
name: "FastConformer-CTC-BPE"
# name: "model STT N tes…
-
We are simultaneously trying minimize SDK size and model size for the cascaded pose regression process. A well tuned gradient boosting regression may often outperform random forests, however, the nee…
-
### Have you searched existing issues? 🔎
- [X] I have searched and found no existing issues
### Desribe the bug
After running `reduce_outliers` and `update_topics`, the effects of all specif…
shj37 updated
3 months ago
-
### Discussed in https://github.com/MaartenGr/BERTopic/discussions/2199
Originally posted by **hari-chalise** October 26, 2024
Hi Maarten, I am trying to Topic modelling BY using BERTopic for …
-
thanks for the great work. I was trying to reproduce your code, I noticed during pretraining, if you set the `mm_vision_output_token_count = 576` you will get:
```
File "llava-token-compression/ll…
-
### Checklist
- [X] I have searched for [similar issues](https://github.com/isl-org/Open3D/issues).
- [X] For Python issues, I have tested with the [latest development wheel](https://www.open3d.org/d…
-
The model estimates 18 age classes. Of those, 0.1, 0.5, 1.5, 3.1, 3.4 are practically zero. In my experience, this tend to lead to model instability.
-
**Describe**
I found that after finetuning with Lora, the token throughput is significantly reduced. I trained a model on the unit test generation. And then fused the Lora adapter.
For my test dat…
-
## タイトル: 大規模視覚言語モデルにおける中間層の悪魔：アテンションレンズを用いた物体幻覚の解釈、検出、および軽減
## リンク: https://arxiv.org/abs/2411.16724
## 概要:
大規模視覚言語モデル (LVLM) の幻覚は信頼性を著しく低下させるため、その原因究明が盛んに行われています。しかし、既存研究の多くは視覚的側面より言語的側面に焦点を当てていま…