alwint3r (Alwin Arrasyid)

liked a model about 1 month ago

Qwen/Qwen2.5-7B-Instruct

Text Generation • Updated Sep 25 • 1.41M • 316

liked a model 4 months ago

cahya/whisper-medium-id

Automatic Speech Recognition • Updated May 22 • 285 • 20

upvoted a paper 4 months ago

DataComp-LM: In search of the next generation of training sets for language models

Paper • 2406.11794 • Published Jun 17 • 49

liked a model 5 months ago

alvdansen/littletinies

Text-to-Image • Updated Jun 16 • 2.54k • • 390

liked a model 6 months ago

microsoft/Phi-3-vision-128k-instruct

Text Generation • Updated Aug 20 • 150k • 933

liked 2 datasets 7 months ago

oscar-corpus/OSCAR-2301

Updated Apr 18, 2023 • 11.3k • 146

oscar-corpus/OSCAR-2201

Updated May 30, 2023 • 1.26k • 114

New activity in mlx-community/dolphin-2.9-llama3-8b-8bit-mlx 7 months ago

Is model.safetensors necessary here?

#1 opened 7 months ago by

alwint3r

liked a model 8 months ago

cognitivecomputations/dolphin-2.8-mistral-7b-v02

Text Generation • Updated May 20 • 1.15k • 213

New activity in cognitivecomputations/dolphin-2.8-mistral-7b-v02 8 months ago

add bos_token to chat template

1

#10 opened 8 months ago by

alwint3r

Reacted to vladbogo's post with 🤯 9 months ago

Post

"LLM Agents can Autonomously Hack Websites" is a new paper that investigates the capacity of LLMs to autonomously execute cybersecurity attacks on websites, such as SQL injections without human guidance.

Key points:
* It uses a LLM integrated with Playwright, a headless web browser, enabling automated web interactions through function calling.
* It gives access to the LLM to 7 web hacking documents and planning capabilities through specific prompting, without disclosing the exact methods to prevent misuse.

GPT-4 achieves a 73.3% success rate on the tested vulnerabilities, emphasizing the potential cybersecurity risks posed by advanced LLMs. Other open models cannot yet perform these types of attacks (results in screenshot).

Congrats to the authors for their work!

Paper: LLM Agents can Autonomously Hack Websites (2402.06664)