-
Thanks for your excellent work!
In Table 3 of your paper, I wonder what is the difference between "**SFT LLM**" and "**+ SFT**"?
![image](https://github.com/Timothy023/RLMEC/assets/70904956/d6d3…
-
**我只能说up主是一个很会讲故事的人。**
起初我对UP主在B站视频里讲的内容比较感兴趣,因为公司有资源,于是在Mac和Linux两个平台上将视频内容都实现了一遍,包括了up github中的默认训练参数,以及其他的一些自己调参。
这里先说几个结论,再说让我很不爽的点。
**结论:**
1. 达到视频中最终效果的主要因素:每个问题前,加上了提示词:”你是一位B站老用户,请使用…
-
在以下文档中没有找到我需要的常规**部署指令**,请教一下
[Vllm推理部署](https://github.com/modelscope/swift/blob/main/docs/source/LLM/VLLM%E6%8E%A8%E7%90%86%E5%8A%A0%E9%80%9F%E4%B8%8E%E9%83%A8%E7%BD%B2.md#%E5%BE%AE%E8%B0%83%E5%90%…
-
hey,
you have no discussion are so here ;)
what is better in RAG than "pdfgear"
https://www.pdfgear.com/de/
its 300MB (offline) and better than all i tryed before ...
how you prevent in …
-
Check how GPT labels statements on our labeling task. Use $Global R^2 = 1-\frac{mse(prediction,actual)}{mse(baseline,actual)}$ to score, and we can visualize in observable.
Would be nice to see ho…
-
### Bug Description
In the LLamaIndex documentation for python, Under Examples section, in data connectors category shows None category, which reroutes to blank page.
![image](https://github.com/run…
-
### Description
I'm trying out Big-AGI with Claude 3.5 Sonnet, and it seems the UI can't handle bulletpoints in markdown.
Inspecting the HTML in the browser, it's just a `` tag and a bunch of bull…
-
Train 3 binary 'Mission Classifiers' i.e one classifier for AHL, AFS, ASF:
- Start with simple keyword matching approach
- Try more advanced methods such as embeddings + classifiers or fine-tuning a…
-
Hi Developers,
I have a naive question, could you please help me understand:
- During the model training, are the weights and biases of the BERT model(or any other supported model) changing? (I…
-
Hello!
In the paper, it is said that Chat-3D-v2 can handle various downstream tasks like 3D QA, 3D Visual grounding, 3D Dense Captioning and 3D Scene captioning. Does it mean that the model can handl…