Sources · Twitter/X

@jxmnop

Active

jxmnop

Linked items 22
Errors 17
Last fetched 2026-05-10 01:40
Priority P3

Edit

Human-friendly label. Leave blank to fall back to the identifier.

Drives ingestion cadence and the default ranking weight applied to items from this source.

How often the watch loop polls this source. Minimum 5 min. Leave blank for the per-type default.

Optional. Pin a specific summarization model for items from this source.

Cancel

Recent items

The 15 most-recently-ingested items from this source, newest first.

Title Type Fetched
@jxmnop: The AI world is in crisis precisely because man has strayed too far from his datacenters. one should... article 2026-05-08
@jxmnop: every ML conference I've been to since 2019 has had no fewer than three papers proposing new techniq... article 2026-05-07
@jxmnop: when my phd advisor asks me for a weekly update on my experiments article 2026-05-07
@jxmnop: "Introducing a breakthrough new technique for sub-quadratic attention, making long-context LLMs 10x ... article 2026-05-07
@jxmnop: people on here are dumb. the latest subquadratic attention trick might produce a model that *process... article 2026-05-06
@jxmnop: "1M context" models after 100k tokens dr. jack morris (@jxmnop) it is endlessly fascinating to ... article 2026-05-05
@jxmnop: Real context compression has never been tried article 2026-05-05
@jxmnop: with Codex, i can run 10x the experiments out of these experiments, i can trust about 15% of the re... article 2026-05-04
@jxmnop: it is endlessly fascinating to me that we still don't have a true 1M-context model it's an unusual ... article 2026-05-04
@jxmnop: the OpenAI goblin fiasco was a Big L for the interpretability research community They solved the my... article 2026-05-02
@jxmnop: very interesting that Claude Code is the ultimate product for vibecoding, and Claude Code's engineer... article 2026-04-29
@jxmnop: over the last few years ~four american AI labs have stockpiling many petabytes of Good Data. no one... article 2026-04-27
@jxmnop: he’s offended because they didn’t benchmax ARC AGI :( François Chollet (@fchollet) The new model ... article 2026-04-08
@jxmnop: it's pretty funny to me that Mythos is capable of hacking entire nation states but it still couldn't... article 2026-04-08
@jxmnop: Hate to break it to you, but the first LLM was created by Andrey Markov in 1913. he tallied up 20,0... article 2026-03-30