C Thi Nguyen
@add_hawk · Salt Lake City, UT · joined 21 Aug 2015
Philosophy professor. Writes about games, trust, art, intimacy, echo chambers, metrics. Also posts @add-hawk.bsky.social
24 235Followers
1 732Following
6 981Posts total
83.5KViews on collected posts
Últimas publicaciones
Buscando publicaciones nuevas…
People will say things like, "This made my day" or week or whatever, but the WaPo review of my book... made me feel like I have not wasted my life.
10.7K views · 112 likes
· 11 reposts · 13 replies
Open on X →
I am also now on BlueSky - @add-hawk.bsky.social - find me and let me find you.
2.5K views · 11 likes
· 1 reposts · 2 replies
Open on X →
The Nightmare Before Christmas as a meditation on cultural appropriation
4.3K views · 28 likes
· 4 reposts · 2 replies
Open on X →
Thank you @add_hawk for this lovely thread! If people want to read more:
- name artifacts paper by @VeredShwartz, @rachelrudinger, & Tafjord: https://t.co/FPNaCYMyM2
- our papers:
https://t.co/QalXhLSbwy
https://t.co/9CS8yUGGev https://t.co/xWFLKgv6DT https://t.co/YC0EvZ1cou
6.5K views · 35 likes
· 4 reposts · 1 replies
Open on X →
@add_hawk You can also see the impact on names in performance on reasoning problems. Change the name, get a different answer.
https://t.co/PWc9yrtmg9
962 views · 8 likes
· 0 reposts · 0 replies
Open on X →
PS @KathleenACreel is a trained philosopher and software engineer who’s appointed as a prof in both phil and CS, and she has proposals for technical solutions to partially ameliorate these problems which are above my pay grade.
6.4K views · 46 likes
· 3 reposts · 2 replies
Open on X →
Anyway, somehow the naming artifacts problem lets me understand, in my gut, what I've already heard many times but never fully processed: exactly the degree to which LLMs are not thinking, but are mass averaging parrot devices.
2.7K views · 62 likes
· 7 reposts · 4 replies
Open on X →
And, she says, this is a tough problem to get out of, because it's hard to train different models on different data sets - because there's only one Internet.
4.5K views · 38 likes
· 0 reposts · 5 replies
Open on X →
@KathleenACreel says: there are a bunch of different corporate products, but they're usually adaptations of one original model. Or even if they aren't, they're different models trained on the same base data set.
2.9K views · 34 likes
· 1 reposts · 1 replies
Open on X →
And she says there's a bunch of distinctive new LLM biases, like that naming artifact. Which is because LLMs are trained on the internet, and the internet is, like, 60% by weight Reddit. So resume algorithms are all systematically biased against anybody named Donald or Hillary.
3.3K views · 48 likes
· 4 reposts · 1 replies
Open on X →
This creates a distinctive new social harm: the possibility that you'll get shut out of *every* job search because your resume doesn't fit that one model.
3.2K views · 43 likes
· 5 reposts · 1 replies
Open on X →
I learned this from @KathleenACreel and her AI ethics work on "the algorithmic monolith". Like when the majority of HR departments use a handful of resume-filtering algorithms for picking who to interview, which are all drawn from one original model.
8.9K views · 41 likes
· 2 reposts · 1 replies
Open on X →
The example that helps me intuitively understand how LLMs "work, more than anything else: "naming artifacts":
It turns out, LLMs are systematically biased against anybody with the following names: Donald, Barack, Hillary, Bernie.
21.3K views · 137 likes
· 20 reposts · 2 replies
Open on X →
What works for "Alice" does not necessarily work for "Jess"
Simple Alice in Wonderland question. Only change is the name. Different answers.
Claude 3.5 Sonnet, temp = 0 (through the console) https://t.co/mDPGLtzNYt
2.5K views · 11 likes
· 0 reposts · 1 replies
Open on X →
My book, GAMES: AGENCY IS ART is out! It's about:
How game designers sculpt agency.
How games let us record, transmit, and explore new forms of agency.
How real games make us more free.
How gamification undermines our freedom. https://t.co/C0Ef0HZLJv
0 views · 1.3K likes
· 210 reposts · 57 replies
Open on X →
Growth & engagement
How the posts we collected actually performed: views and reaction rate post by post, what the audience did with them, and where the follower count goes.
Views per post
21.3K2 Jul
8.9K
3.2K
3.3K
2.9K
4.5K
2.7K
6.4K
962
6.5K8 Jul
2.8K10 Jul
4.3K25 Dec
2.5K14 Jul
10.7K12 Jan
Last 14 collected posts, oldest on the left. The scale is logarithmic: one post can outrun the rest a hundred times over.
Engagement rate per post
0.77%2 Jul
0.51%
1.53%
1.65%
1.25%
0.95%
2.66%
0.80%
0.83%
0.63%8 Jul
0.11%10 Jul
0.80%25 Dec
0.56%14 Jul
1.27%12 Jan
Reactions — likes, reposts, replies and quotes — divided by views.
What the audience does
Likes74.7%657 in total
Reposts7.0%62 in total
Replies4.1%36 in total
Quotes1.0%9 in total
Bookmarks13.2%116 in total
Share of every reaction we collected for this account. Replies mean argument, reposts mean endorsement, bookmarks mean the post was worth keeping.
The follower curve appears once this account has two daily snapshots — we take one a day, and this one is on its first.
Cuentas similares
山本一成🌤️完全自動運転AIを作るチューリングのCEO ✓
@issei_y
48.5K followers
· 38.2K posts
𝙋𝘼𝙐𝙇 𝐍𝐄𝐊𝐎 ✓
@ouma_neko
48.5K followers
· 24.1K posts
Mohammed
@FCIM_Pictures
48.5K followers
· 6.3K posts
ludolullabi
@ludolullabi
48.5K followers
· 4.7K posts
Carlin Correa
@carlincorrea
48.5K followers
· 72K posts
Fred Bauma
@fredbauma90
48.5K followers
· 18.1K posts
Bernarda Llorente ✓
@bernardallorent
48.5K followers
· 11.8K posts
ミニぜ
@minize_55
48.5K followers
· 93 posts