RT Alexander DoriaAnd new data release: in partnership with @GSMA we publish Telco-Common-Corpus, the largest fully open corpus for the telecom sector, 10 billion tokens all…
> However, these new checkpoints are weird. They mention March 2026 knowledge cutoff but they don't know major stuff from late 2025.So. mid-2025. Gemini has recovered…
I don't like the results so now the benchmark sucks.LONGTERMONLY: @teortaxesTex You don't like the results so now the benchmark sucks?
Do the Japanese stereotype Chinese beauties as "tall, small boobs, long legs"? And they're saying this as if it's something bad?ミス東大2026 No.5 曾庭榕: こんばんは🌙実験の休憩中の私と、ゴルフをしている私⛳️みなさんは普段、どうやってストレス発散していますか？😊私は時間があると友達とゴルフに行って、自然の中でリフレッシュしています🍃でも、なかなか上達しません…😂#東京大学#ミスコン
GLM 5.2 is the best Chinese model on ARC-AGI-2, at 22.8% (is that high or max?), on par with Opus 4.5 (16K). …Whereas Grok 4.20 is…
Zyphra fits a scaling law for plasticity loss in continuously trained LLMs. What can we do to push the point of rigidity onset towards infinity? I…
one of the two must stay in the lockereither Fable, or DarioDario is a weirdo, so Fable gets to walkTeortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞): Life…
I think Cheng is seriously trying to avoid a disaster. If it were just about Taiwanese independence and CCP's long term objectives, Xi would certainly tolerate…
R1 is unparalleled to this dayWe should try to revive it in V4/V5Elyas Masrour: Did you know?Pangram learns the difference between Claude, ChatGPT, and Gemini in…
…depressingTrump is like a fish that keeps going after the bait (same exact bait) after losing one eye. Since the Liberation DayZephyr: Looks like the next…