10/ LLM Agents can Hack – shows that LLM agents can automatically hack websites and perform tasks like SQL injections without human feedback or explicit knowledge about the vulnerability beforehand.
AI
-

TestGen-LLM Uses AI to Improve Software Testing Coverage
By
–
7/ TestGen-LLM – uses LLMs to improve existing human-written tests; reports that after an evaluation on Reels and Stories products for Instagram, 75% of TestGen-LLM's test cases were built correctly, 57% passed reliably, and 25% increased coverage.
-

OS-Copilot Framework Builds Generalist Computer Agents
By
–
6/ OS-Copilot – a framework to build generalist computer agents that interface with key elements of an operating system like Linux or MacOS; it also proposes a self-improving embodied agent for automating general computer tasks.
-

Gemini 1.5: Compute-Efficient Multimodal Mixture-of-Experts Model
By
–
2/ Gemini 1.5 – a compute-efficient multimodal mixture-of-experts model with capabilities such as recalling and reasoning over long-form content; reasons over long documents potentially containing millions of tokens, including hours of video and audio.https://t.co/oZyOtMNBnR
— DAIR.AI (@dair_ai) 18 février 20242/ Gemini 1.5 – a compute-efficient multimodal mixture-of-experts model with capabilities such as recalling and reasoning over long-form content; reasons over long documents potentially containing millions of tokens, including hours of video and audio.
-

V-JEPA: Self-Supervised Vision Model Trained on Two Million Videos
By
–
3/ V-JEPA – vision models trained on a feature prediction objective using 2 million videos; relies on self-supervised learning and doesn’t use pretrained image encoders, text, negative examples, reconstruction, or other supervision sources.
-
Top ML Papers Week: Sora, Gemini 1.5, Agents
By
–
The Top ML Papers of the Week (Feb 12 – Feb 18): – Sora
– Gemini 1.5
– OS-Copilot
– TestGen-LLM
– Large World Model
– LLM Agents can Hack
… -
Sora: Text-to-Video AI Model Generates Realistic Minute-Long Videos
By
–
1/ Sora – a text-to-video AI model that can create videos of up to a minute of realistic and imaginative scenes given text instructions; it can generate complex scenes with multiple characters, different motion types, and backgrounds, and understand how they relate to each other.
-
Web3 Metaverse Strategy for Binance Coins and Memes
By
–
Can I make a suggestion for @Web3WithBinance
? Given UX, tech & new trends like #brc20 are in place. Can you use the #Web3 narrative to create a #metaverse where coins/ #memes can unite? Work, earn, spend & game. It could be a new #USP. @_RichardTeng @heyibinance @RachelConlan -
Automating Account Data via API Would Be Useful
By
–
Well, what would really be useful is automating account data via API like with Copilot. That's what's needed.
-

E/accs advocates for acceleration and AI promotion
By
–
E/accs need to promote the use of acceleration and ai.
