Tim
@timzyzz.bsky.social
I'm a researcher interested in how the internet mediates cultural evolution
Working on recommender systems for "societal good" (whatever that means)
timbooker.net
Working on recommender systems for "societal good" (whatever that means)
timbooker.net
So, Albanese is upset with waiting times? Welcome to Medicare 😀
New: Another OpenAI agent hacked the Australian government.
PM says he expressed his disappointment "that it took [OpenAI] way too long to inform the government what had occurred, and the nature of the way that that notification occurred as well was unacceptable."
www.theage.com.au/politics/fed...
PM says he expressed his disappointment "that it took [OpenAI] way too long to inform the government what had occurred, and the nature of the way that that notification occurred as well was unacceptable."
www.theage.com.au/politics/fed...
OpenAI agent breached Medicare, Albanese reveals
An artificial intelligence agent developed by OpenAI hacked Medicare this year, in a breach that Prime Minister Anthony Albanese has labelled “unacceptable”.
www.theage.com.au
September 24, 2026 at 11:22 AM
So, Albanese is upset with waiting times? Welcome to Medicare 😀
Reposted by Tim
Very difficult to draw lines that simultaneously make humans special and also let you eat all the meat you want and not give a shit about AI
September 15, 2026 at 3:24 PM
Very difficult to draw lines that simultaneously make humans special and also let you eat all the meat you want and not give a shit about AI
announcement: i am permanently banned from chatgpt without possibility of appeal
September 14, 2026 at 12:54 PM
announcement: i am permanently banned from chatgpt without possibility of appeal
Reposted by Tim
the current “anti-AI” debate is like getting hit by a nuclear bomb and complaining that you don’t like the hue of the mushroom cloud
September 3, 2026 at 12:04 PM
the current “anti-AI” debate is like getting hit by a nuclear bomb and complaining that you don’t like the hue of the mushroom cloud
Reposted by Tim
🤖🧠NEW PAPER🧠🤖
(The result of an 8-year project!)
LLMs seem very different from symbolic systems. Yet LLMs excel in symbolic domains (e.g., language/code/math). How do they do it?
Our finding: LLM representations have implicit symbolic structure!
Link in thread ⬇️
1/n
(The result of an 8-year project!)
LLMs seem very different from symbolic systems. Yet LLMs excel in symbolic domains (e.g., language/code/math). How do they do it?
Our finding: LLM representations have implicit symbolic structure!
Link in thread ⬇️
1/n
![Overview of the paper.
Title: The Emergent Symbolic Structure of Artificial Neural Networks
Authors: Tom McCoy, Paul Soulos, Tal Linzen, Paul Smolensky
Left: Neural networks encode information in vectors (there is then an image of a vector), yet they excel at tasks long thought to require symbolic structure (there is then an image of a symbolic representation, specifically a syntax tree). How do LLMs do it?
Right: We find that LLM representations can be closely approximated with symbolic structures. This approximation lets us edit the structure of an LLM’s output by editing the structure of its internal representations, as shown. There is then an image of two edits to LLMs. In the first one, the original input is 3 + 6 * 8, with an answer of 51. But if we swap the positions of the 3 and the 6, the output becomes 30. In the second one, the original input is a Python command repeating the list [Z, U] three times, producing [Z, U, Z, U, Z, U]. But if we edit the input in a way that adds a Q at the end of the input, the output becomes [Z, U, Q, Z, U, Q, Z, U, Q].](https://cdn.bsky.app/img/feed_thumbnail/plain/did:plc:bcsvlv7ubeagj5virdvatadd/bafkreic3zm2rpc4vtqrz2t3y4vcl3t2owp2sxay42vuh5zm5zkg4jhr64e)
September 1, 2026 at 7:21 PM
🤖🧠NEW PAPER🧠🤖
(The result of an 8-year project!)
LLMs seem very different from symbolic systems. Yet LLMs excel in symbolic domains (e.g., language/code/math). How do they do it?
Our finding: LLM representations have implicit symbolic structure!
Link in thread ⬇️
1/n
(The result of an 8-year project!)
LLMs seem very different from symbolic systems. Yet LLMs excel in symbolic domains (e.g., language/code/math). How do they do it?
Our finding: LLM representations have implicit symbolic structure!
Link in thread ⬇️
1/n
Looking forward to this fuckin flight to Helsinki leesgoo #rebuild @ruggsea.eurosky.social @lorrenz.bsky.social
August 29, 2026 at 9:20 AM
Looking forward to this fuckin flight to Helsinki leesgoo #rebuild @ruggsea.eurosky.social @lorrenz.bsky.social
Reposted by Tim
Esteemed author Richard Flanagan delivered a powerful speech at the Festival of Dangerous Ideas warning of a new McCarthyism, accusing Israel of "seeking to silence" Australians with the slur of antisemitism.
We have published the full text of the speech.👇️
www.deepcutnews.com/p/segals-pla...
We have published the full text of the speech.👇️
www.deepcutnews.com/p/segals-pla...
The new McCarthyism: Gaza and Australia – Richard Flanagan
This is the full text of a powerful speech Flanagan delivered at the the Festival of Dangerous Ideas on August 22
www.deepcutnews.com
August 25, 2026 at 9:12 PM
Esteemed author Richard Flanagan delivered a powerful speech at the Festival of Dangerous Ideas warning of a new McCarthyism, accusing Israel of "seeking to silence" Australians with the slur of antisemitism.
We have published the full text of the speech.👇️
www.deepcutnews.com/p/segals-pla...
We have published the full text of the speech.👇️
www.deepcutnews.com/p/segals-pla...
Reposted by Tim
Check out the largest open multilingual preprocessed news corpus from Common Crawl News (built with @ruggsea.eurosky.social & @janalasser.eurosky.social).
1.36B articles | subsecond search | FOSS
Demo: infini-news.uni-graz.at
Paper: arxiv.org/abs/2605.18337
Code: codeberg.org/ksolovev/inf...
1.36B articles | subsecond search | FOSS
Demo: infini-news.uni-graz.at
Paper: arxiv.org/abs/2605.18337
Code: codeberg.org/ksolovev/inf...
infini-news
infini-news: search 1.36B+ news articles with sub-second full-text search
infini-news.uni-graz.at
July 1, 2026 at 12:06 PM
Check out the largest open multilingual preprocessed news corpus from Common Crawl News (built with @ruggsea.eurosky.social & @janalasser.eurosky.social).
1.36B articles | subsecond search | FOSS
Demo: infini-news.uni-graz.at
Paper: arxiv.org/abs/2605.18337
Code: codeberg.org/ksolovev/inf...
1.36B articles | subsecond search | FOSS
Demo: infini-news.uni-graz.at
Paper: arxiv.org/abs/2605.18337
Code: codeberg.org/ksolovev/inf...
Reposted by Tim
Here's my new blogpost on the dataset I have created alongside @ksolovev.com: we decided to solve the issue of data availability for research on news articles by processing, cleaning, tagging and indexing the whole news subset of the Common Crawl, resulting in
cs2.uni-graz.at/blog/infini-...
cs2.uni-graz.at/blog/infini-...
July 1, 2026 at 12:16 PM
Here's my new blogpost on the dataset I have created alongside @ksolovev.com: we decided to solve the issue of data availability for research on news articles by processing, cleaning, tagging and indexing the whole news subset of the Common Crawl, resulting in
cs2.uni-graz.at/blog/infini-...
cs2.uni-graz.at/blog/infini-...
Reposted by Tim
A friend and colleague at my lab (the CS² under @janalasser.bsky.social ) is organizing a super nice school on democracy and social media here in Graz!
If you're interested, check it out and share! Link: sicss.io/2025/graz/
If you're interested, check it out and share! Link: sicss.io/2025/graz/
March 27, 2025 at 2:46 PM
A friend and colleague at my lab (the CS² under @janalasser.bsky.social ) is organizing a super nice school on democracy and social media here in Graz!
If you're interested, check it out and share! Link: sicss.io/2025/graz/
If you're interested, check it out and share! Link: sicss.io/2025/graz/