BenchMAX: A Comprehensive Multilingual Evaluation Suite for Large Language Models
Paper
•
2502.07346
•
Published
•
44
FeiYuan
FeYuan
AI & ML interests
None yet
Recent Activity
upvoted
a
paper
4 days ago
BenchMAX: A Comprehensive Multilingual Evaluation Suite for Large
Language Models
upvoted
a
paper
3 months ago
OS-ATLAS: A Foundation Action Model for Generalist GUI Agents
Organizations
FeYuan's activity
upvoted
a
paper
4 days ago
upvoted
a
paper
3 months ago
upvoted
a
paper
4 months ago
authored
2
papers
5 months ago
upvoted
a
paper
5 months ago
commented
a
paper
5 months ago
liked
a
Space
7 months ago
replied to
their
post
7 months ago
Thanks for your helpful suggestion and your interest in LLaMAX.
As per your suggestion, we have tagged LLaMAX as multilingual on HF.
upvoted
a
paper
7 months ago
authored
a
paper
7 months ago
posted
an
update
7 months ago
Post
4781
Hi everyone,
I am excited to introduce our latest work, LLaMAX. 😁😁😁
LLaMAX is a powerful language model created specifically for multilingual scenarios. Built upon Meta's LLaMA series models, LLaMAX undergoes extensive training across more than 100 languages.
Remarkably, it enhances its multilingual capabilities without compromising its generalization ability, surpassing existing LLMs.
✨Highlights:
🎈 LLaMAX supports the 102 languages covered by Flores-101, and its performance in translating between low-resource languages far surpasses other decoder-only LLMs.
🎈 Even for languages not covered in Flores-200, LLaMAX still shows significant improvements in translation performance.
🎈 By performing simple SFT on English task data, LLaMAX demonstrates impressive multilingual transfer abilities in downstream tasks.
🎈 In our paper, we discuss effective methods for enhancing the multilingual capabilities of LLMs during the continued training phase.
We welcome you to use our model and provide feedback.
More Details:
🎉 Code: https://github.com/CONE-MT/LLaMAX/
🎉 Model: https://huggingface.co/LLaMAX/
I am excited to introduce our latest work, LLaMAX. 😁😁😁
LLaMAX is a powerful language model created specifically for multilingual scenarios. Built upon Meta's LLaMA series models, LLaMAX undergoes extensive training across more than 100 languages.
Remarkably, it enhances its multilingual capabilities without compromising its generalization ability, surpassing existing LLMs.
✨Highlights:
🎈 LLaMAX supports the 102 languages covered by Flores-101, and its performance in translating between low-resource languages far surpasses other decoder-only LLMs.
🎈 Even for languages not covered in Flores-200, LLaMAX still shows significant improvements in translation performance.
🎈 By performing simple SFT on English task data, LLaMAX demonstrates impressive multilingual transfer abilities in downstream tasks.
🎈 In our paper, we discuss effective methods for enhancing the multilingual capabilities of LLMs during the continued training phase.
We welcome you to use our model and provide feedback.
More Details:
🎉 Code: https://github.com/CONE-MT/LLaMAX/
🎉 Model: https://huggingface.co/LLaMAX/
-
- 3 replies
upvoted
a
paper
7 months ago
commented
a
paper
7 months ago
reacted to
Symbol-LLM's
post with 🚀👍
11 months ago
Post
1744
Check out our work Symbol-LLM ! We have open-sourced both 7B / 13B model weights, as well as part of the symbolic collections ! Try it !
Paper link: Symbol-LLM: Towards Foundational Symbol-centric Interface For Large Language Models (2311.09278)
Model weights: Symbol-LLM/Symbol-LLM-7B-Instruct
Paper link: Symbol-LLM: Towards Foundational Symbol-centric Interface For Large Language Models (2311.09278)
Model weights: Symbol-LLM/Symbol-LLM-7B-Instruct
authored
4
papers
11 months ago
Scaling TransNormer to 175 Billion Parameters
Paper
•
2307.14995
•
Published
•
22
Extrapolating Large Language Models to Non-English by Aligning Languages
Paper
•
2308.04948
•
Published
•
1
Symbol-LLM: Towards Foundational Symbol-centric Interface For Large Language Models
Paper
•
2311.09278
•
Published
•
7
How Multilingual is Multilingual LLM?
Paper
•
2311.09071
•
Published