Top Tweets for #workarena
Surprising results of GPT-5 on #WorkArena
๐จ Is #WorkArena on the verge of being solved? Or did GPT-5 just get trained on it?
๐ฅWhile some benchmarks show modest gains, GPT-5 is crushing WorkArena L2๐ฅ
โก๏ธ 69.4% avg success vs. ~40% for next best๐คฏ
โก๏ธ Complex tasks, up to 100 steps, 5โ20 min for humans
๐จ Is #WorkArena on the verge of being solved? Or did GPT-5 just get trained on it?
๐ฅWhile some benchmarks show modest gains, GPT-5 is crushing WorkArena L2๐ฅ
โก๏ธ 69.4% avg success vs. ~40% for next best๐คฏ
โก๏ธ Complex tasks, up to 100 steps, 5โ20 min for humans
@ysu_nlp Very cool results!
I'm surprised about Claude Computer use. When We tried Claude-3.5 on #WorkArena, it was by far the best model.
https://t.co/bLtS6hqEuR
@AIatMeta ๐ฅ Fresh off the GPU, new UI WebAgent results on #WorkArena-L1 are in! ๐ท Llama 3.3 70B: 34.5% (โ6.6% from 3.1) Qwen 2.5 32B: 27.9% Even the small models shine: Qwen 2.5 7B (8.2%) doubles Llama 3.1 8B (4%)! ๐ทโ๏ธ
Notable findings:
๐Claude-3.5-Sonnet is insanely good on #WorkArena L2
๐ชจ #WorkArena L3 is insanely hard
๐คo1-mini is quite good across many benchmarks
๐ฒo1 is very expensive :)
๐ Internship Opportunity on web agents at ServiceNow Research! ๐
We aim to boost ๐ the performance in #WorkArena and #WebArena by fine-tuning large open-source LLMs.
(Candidate must be affiliated with a Canadian university)
๐ Apply Here: https://t.co/pDO8xauniy
Hurray ๐!
Our #WorkArena / #BrowserGym paper is accepted to ICML.
See you in Vienna โ๏ธ
https://t.co/40v6dZCYyH
How capable are web agents at solving knowledge work tasks? ๐ค Are LLMs up to the challenge? ๐ค
Introducing WorkArena: a benchmark where agents meet the world ๐ธ๐ช๐ญ๐ฅ web of enterprise software ๐๐ฅ๏ธ
Paper: https://t.co/9WvM7xOWyC
Website: https://t.co/lfNj3LnzmV
๐งต 1/7
๐งต2/4
paper: https://t.co/TjvvuEyMPa
The original aim was to build a strong backend for our new benchmark #WorkArena. Since our backend is flexible we included MiniWoB++ and WebArena.
Part of the gain comes from a better extraction of the accessibility tree.
Very excited about #Llama3 . We are finally able to get some tasks to be solved in #WorkArena using open-source LLMs. Previous attempts always gave 0% except for GPT.
https://t.co/hdO9P4XSMo
๐ LLAMA3 is the first open-source LLM to ace tasks in Workarena, making it the top OSS virtual knowledge worker! (and believe us we've tested many models and prompting techniques ๐
) Watch it excel in a challenging knowledge base task. Kudos to @AIatMeta for the amazing model ๐
& THE WINNER IS @avathyrza! You are the winner of our #NAWICNSWAwards IG Comp. Enjoy the Eames Chairs from #Workarena
Last Seen Hashtags on Sotwe
Most Popular Users
Elon Musk 
@elonmusk
241.7M followers
Barack Obama
@barackobama
119M followers
Cristiano Ronaldo
@cristiano
114.5M followers
Donald J. Trump
@realdonaldtrump
111.9M followers
Narendra Modi
@narendramodi
107.2M followers
Rihanna
@rihanna
98.7M followers
NASA
@nasa
92.4M followers
Justin Bieber
@justinbieber
91.8M followers
KATY PERRY
@katyperry
90M followers
Taylor Swift
@taylorswift13
84M followers
Lady Gaga
@ladygaga
75.5M followers
Virat Kohli
@imvkohli
73.5M followers
Kim Kardashian
@kimkardashian
70.9M followers
YouTube
@youtube
68.8M followers
Neymar Jr
@neymarjr
66.5M followers
Bill Gates
@billgates
65.2M followers
Selena Gomez
@selenagomez
63.1M followers
The Ellen Show
@theellenshow
62.3M followers
CNN
@cnn
61.8M followers
X
@x
60.7M followers