Haebin Seong

hbseong

AI & ML interests

None yet

Recent Activity

Organizations

None yet

hbseong's activity

posted an update 20 days ago
view post
Post
724
šŸšØšŸ”„ HOT MODEL accepted at ICLR!! Currently 2137 downloads!!! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out:
hbseong/HarmAug-Guard
(Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
posted an update 3 months ago
view post
Post
928
šŸšØšŸ”„ HOT MODEL!! Currently 870 downloads!!! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
  • 1 reply
Ā·
posted an update 3 months ago
view post
Post
930
šŸšØšŸ”„ New Release Alert! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
  • 1 reply
Ā·
posted an update 3 months ago
view post
Post
3304
šŸšØšŸ”„ New Release Alert! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
reacted to their post with ā¤ļøšŸš€ 4 months ago
view post
Post
1194
šŸšØšŸ”„ New Release Alert! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
posted an update 4 months ago
view post
Post
1194
šŸšØšŸ”„ New Release Alert! šŸ”„šŸšØ

Introducing the 435M model that outperforms Llama-Guard-3-8B while slashing 75% of the computation cost! šŸ’»šŸ’„
šŸ‘‰ Check it out: hbseong/HarmAug-Guard (Yes, INFERENCE CODE INCLUDED! šŸ’”)

More details in our paper: https://arxiv.org/abs/2410.01524 šŸ“œ

#HarmAug #LLM # Safety #EfficiencyBoost #Research #AI #MachineLearning
New activity in hbseong/HarmAug-Guard 4 months ago

Upload README.md

#3 opened 4 months ago by
Seanie-lee
New activity in hbseong/HarmAug-Guard 4 months ago

Upload README.md

#2 opened 4 months ago by
Seanie-lee
New activity in hbseong/HarmAug-Guard 4 months ago

add eval() and no_grad()

#1 opened 4 months ago by
Seanie-lee
updated a collection over 1 year ago