Dr Sarah
@drsezzer.bsky.social
200 followers
140 following
440 posts
Software Engineer / GenAI researcher (LLM agents) ex-The Alan Turing Institute. Inbetween jobs.
Witcher fan (books/games not netflix). Recreates retro games in python for fun! *Opinions are mine* or borrowed from those more insightful.
Posts
Media
Videos
Starter Packs
Pinned
Dr Sarah
@drsezzer.bsky.social
· Feb 5
Reposted by Dr Sarah
Reposted by Dr Sarah
Can AI simulations of human research participants advance cognitive science? In @cp-trendscognsci.bsky.social, @lmesseri.bsky.social & I analyze this vision. We show how “AI Surrogates” entrench practices that limit the generalizability of cognitive science while aspiring to do the opposite. 1/
AI Surrogates and illusions of generalizability in cognitive science
Recent advances in artificial intelligence (AI) have generated enthusiasm for using AI simulations of human research participants to generate new know…
www.sciencedirect.com
Reposted by Dr Sarah
The Register
@theregister.com
· 10d
Tech industry grad hiring crashes 46% as bots do junior work
GenAI meets Gen Z – only one gets the job
ai-pocalypse The UK tech sector is cutting graduate jobs dramatically – down 46 percent in the past year, with another 53 percent drop projected, according to figures from the Institute of Student Employers (ISE).…
dlvr.it
Reposted by Dr Sarah
Jesus christ. AI coding platform Augment code had to jack up prices because 22.5% of their users were spending 20x of what they paid, and even after raising the prices, they're still running at a loss.
reddit.com/r/AugmentCod...
reddit.com/r/AugmentCod...
OpenAI GPT-5: great taste, less filling, now with 30% less bias
OpenAI GPT-5: great taste, less filling, now with 30% less bias
AI model maker touts effort to depoliticize its product
OpenAI says GPT-5 has 30 percent less political bias than its prior AI models.…
dlvr.it
Reposted by Dr Sarah
The blog linked from here is pretty cute. Interesting way to check what models may have been trained on.
Defying Transformers: Searching for "Fixed Points" of Pretrained LLMs by Jiacheng Liu
He wondered what CAN'T be transformed by Transformers? So, he wrote a fun blog post on finding "fixed points" of your LLMs. If you prompt it with a fixed point token,
He wondered what CAN'T be transformed by Transformers? So, he wrote a fun blog post on finding "fixed points" of your LLMs. If you prompt it with a fixed point token,
Defying Transformers: Searching for "Fixed Points" of Pretrained LLMs by Jiacheng Liu
He wondered what CAN'T be transformed by Transformers? So, he wrote a fun blog post on finding "fixed points" of your LLMs. If you prompt it with a fixed point token,
He wondered what CAN'T be transformed by Transformers? So, he wrote a fun blog post on finding "fixed points" of your LLMs. If you prompt it with a fixed point token,
