✦Lessy✦

2.9K posts

✦Lessy✦ banner
✦Lessy✦

✦Lessy✦

@lessyfreaky

Like hoot, man | 22 | priv: @lessyfreakier

NYC 🗽 Katılım Mart 2023
640 Takip Edilen872 Takipçiler
Sabitlenmiş Tweet
✦Lessy✦
✦Lessy✦@lessyfreaky·
A piece celebrating my new home in NYC Thank you so much @ashtungg for doing it justice!
✦Lessy✦ tweet media
English
5
13
123
4.8K
✦Lessy✦ retweetledi
Guin
Guin@lance_1_·
おはよう
Guin tweet media
日本語
7
217
3.7K
45.1K
✦Lessy✦ retweetledi
CRITTER
CRITTER@CritterDez·
art trade with @tlactl
CRITTER tweet media
English
0
44
303
2.1K
✦Lessy✦ retweetledi
Pop Base
Pop Base@PopBase·
Today is World Bear Day.
English
668
2.9K
17.8K
7.2M
✦Lessy✦ retweetledi
horse dentist
horse dentist@equine__dentist·
being an adult just means getting really into birds
horse dentist tweet media
English
35
951
12.7K
230.8K
Kazan
Kazan@KazanIsFatzan·
Damn this has got to have been the most productive Saturday I've had in ages. Finally built a new shelf for my small but slowly growing board game collection, cleaned my ice dispenser, hung/moved around a couple pictures, and put up aaaaall my con badges to my pin board.
Kazan tweet mediaKazan tweet media
English
6
0
49
1.5K
✦Lessy✦
✦Lessy✦@lessyfreaky·
Getting wine drunk in Spain fixed my mental more than therapy ever could
English
0
0
3
87
✦Lessy✦ retweetledi
StockBlocc
StockBlocc@Stockblocc·
whoops sorry for dying again, got nothing going on right now so i'll be doing some sketch requests :> feel free to drop your ocs or any other character below
English
15
6
12
504
✦Lessy✦
✦Lessy✦@lessyfreaky·
Is there an owl gc
English
0
0
3
125
Blush🔜FTM
Blush🔜FTM@blushdog_·
Every city should look like this, to live here would be a dream for me
Blush🔜FTM tweet media
English
1
0
11
384
✦Lessy✦ retweetledi
François Chollet
François Chollet@fchollet·
This is more evidence that current frontier models remain completely reliant on content-level memorization, as opposed to higher-level generalizable knowledge (such as metalearning knowledge, problem-solving strategies...)
Lossfunk@lossfunk

🚨 Shocking: Frontier LLMs score 85-95% on standard coding benchmarks. We gave them equivalent problems in languages they couldn't have memorized. They collapsed to 0-11%. Presenting EsoLang-Bench. Accepted to the Logical Reasoning and ICBINB workshops at ICLR 2026 🧵

English
192
329
3K
306K
✦Lessy✦
✦Lessy✦@lessyfreaky·
I am a bird because I am pure of heart
English
0
0
3
37
🤎Bug🤎
🤎Bug🤎@IsoPuppi·
Custom
🤎Bug🤎 tweet media
English
6
255
3.3K
27.8K
✦Lessy✦ retweetledi
Miso
Miso@karasumisso·
He saved you a spot ;}
Miso tweet media
English
10
229
3.3K
26.1K