aqweteddy commited on
Commit
e846079
1 Parent(s): acd179d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +5 -0
README.md CHANGED
@@ -9,6 +9,11 @@ language:
9
  CHAT VECTOR: A SIMPLE APPROACH TO EQUIP LLMS WITH NEW LANGUAGE CHAT CAPABILITIES
10
  https://arxiv.org/pdf/2310.04799.pdf
11
 
 
 
 
 
 
12
  ## Merged LM
13
 
14
  * xwin7b
 
9
  CHAT VECTOR: A SIMPLE APPROACH TO EQUIP LLMS WITH NEW LANGUAGE CHAT CAPABILITIES
10
  https://arxiv.org/pdf/2310.04799.pdf
11
 
12
+
13
+ With the advancements in conversational AI, such as ChatGPT, this paper focuses on exploring developing Large Language Models (LLMs) for non-English languages, especially emphasizing alignment with human preferences. We introduce a computationally efficient method, leveraging “chat vector,” to synergize pre-existing knowledge and behaviors in LLMs, restructuring the conventional training paradigm from continual pretrain
14
+ SFT
15
+ RLHF to continual pretrain + chat. Our empirical studies, primarily focused on Traditional Chinese, employ LLaMA2 as the base model and acquire the chat vector by subtracting the pre-trained weights, LLaMA2, from the weights of LLaMA2-chat. Evaluating from three distinct facets, which are toxicity, ability of instruction following and multi-turn dialogue demonstrates the chat vector's superior efficacy in “chatting”. To confirm the adaptability of our approach, we extend our experiments to include models pre-trained in both Korean and Simplified Chinese, illustrating the versatility of our methodology. Overall, we present a significant solution in aligning LLMs with human preferences efficiently across various languages, accomplished by the chat vector.
16
+
17
  ## Merged LM
18
 
19
  * xwin7b