korrents

On the map

Tap a claim on the ring to put it at the centre.

← Document similarity scores across a corpus tend to cluster bimodally…

17 connected korrents · 15 moments from 1 Mar 2007 to 27 Oct 2026.

Everything filed under documentation documentation Everything filed under LLMs LLMs Everything filed under AI writing AI writing Same subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subjectSame subject Read this korrent: Document similarity scores across a corpus tend to cluster bimodally near identical and near unrelated, rather than spreading evenly. Document similarity scores across a corpustend to cluster bimodally near identicaland near unrelated, rather than spreadingevenly. Last stated 2 years ago 3 Jul 2024 NE Nelson Elhage — holds since 2024-07-03 — tap for who they are Same subject: Judging one model worse than another from a one-shot, probabilistic output on a given day is not a good way to evaluate it. — tap to centre the map on it Judging one model worse than anotherfrom a one-shot, probabilisticoutput on a given day is not a goodway to evaluate it. Last stated 6 months ago 19 May 2026 TG Tommy Geoco — holds since 2026-05-19 — tap for who they are Same subject: The literature may be getting overwhelmed by papers generated by AI. — tap to centre the map on it The literature may be gettingoverwhelmed by papers generated byAI. Last stated a week ago 27 Oct 2026 PW Peter Woit — holds since 2026-10-27 — tap for who they are Same subject: The next-token language modeling objective is misaligned with following user instructions helpfully and safely. — tap to centre the map on it The next-token language modelingobjective is misaligned withfollowing user instructionshelpfully and safely. Last stated 5 years ago 4 Mar 2022 JL Jan Leike — holds since 2022-03-04 — tap for who they are Same subject: A neural network's latent space is closer to an uncopyrightable syntax than to copyrightable expression. — tap to centre the map on it A neural network's latent space iscloser to an uncopyrightable syntaxthan to copyrightable expression. Last stated 2 months ago 7 Sept 2026 KK Kevin Kelly — holds since 2026-09-07 — tap for who they are Same subject: Current techniques restrict pre-trained representation power because standard language models are unidirectional. — tap to centre the map on it Current techniques restrictpre-trained representation powerbecause standard language models areunidirectional. Last stated 8 years ago 11 Oct 2018 KT Kristina Toutanova — holds since 2018-10-11 — tap for who they are MC Ming-Wei Chang — holds since 2018-10-11 — tap for who they are KL Kenton Lee — holds since 2018-10-11 — tap for who they are JD Jacob Devlin — holds since 2018-10-11 — tap for who they are Same subject: Vector embeddings will not solve search: a decades-old term-frequency algorithm still beats most of them at ranking. — tap to centre the map on it Vector embeddings will not solvesearch: a decades-old term-frequencyalgorithm still beats most of themat ranking. Last stated 2 years ago 19 Jun 2024 AS Aravind Srinivas — holds since 2024-06-19 — tap for who they are Same subject: Once a language model reads the results, search can trade precision for recall, because the model does not care that the right link came ninth. — tap to centre the map on it Once a language model reads theresults, search can trade precisionfor recall, because the model doesnot care that the right link cameninth. Last stated 2 years ago 19 Jun 2024 AS Aravind Srinivas — holds since 2024-06-19 — tap for who they are Same subject: Because large language models behave chaotically, a single overall accuracy score cannot capture how reliable they are. — tap to centre the map on it Because large language models behavechaotically, a single overallaccuracy score cannot capture howreliable they are. Last stated 7 months ago 15 Apr 2026 KK Kyle Kingsbury — holds since 2026-04-15 — tap for who they are Same subject: A book as a physical object has no aura, but its text does. — tap to centre the map on it A book as a physical object has noaura, but its text does. Last stated 6 months ago 16 May 2026 VC Vajra Chandrasekera — holds since 2026-05-16 — tap for who they are Same subject: A company that deviates from open standards in its data exports without documenting the deviation creates an artificial vendor lock-in. — tap to centre the map on it A company that deviates from openstandards in its data exportswithout documenting the deviationcreates an artificial vendorlock-in. Last stated 2 years ago 3 Mar 2025 TS Tobias Sjösten — holds since 2025-03-03 — tap for who they are Same subject: A design proposal should describe the feature as though it already existed — once as a tutorial and once as reference documentation. — tap to centre the map on it A design proposal should describethe feature as though it alreadyexisted — once as a tutorial andonce as reference documentation. Last stated 6 months ago 20 May 2026 AR Alice Ryhl — holds since 2026-05-20 — tap for who they are Same subject: A design system is only as good as its documentation. — tap to centre the map on it A design system is only as good asits documentation. Last stated 10 years ago 1 Sept 2016 MM Mina Markham — holds since 2016-09-01 — tap for who they are Same subject: A working specification deliberately resolves ambiguity so that design work can begin. — tap to centre the map on it A working specification deliberatelyresolves ambiguity so that designwork can begin. Last stated 5 months ago 25 May 2026 RR Richard Rumelt — holds since 2026-05-25 — tap for who they are Same subject: Context-sensitive FAQs, done right, can greatly reduce a software user’s cognitive load. — tap to centre the map on it Context-sensitive FAQs, done right,can greatly reduce a software user’scognitive load. Last stated 20 years ago 1 Mar 2007 KS Kathy Sierra — holds since 2007-03-01 — tap for who they are Same subject: A great memo cannot be written in a day or two: it is written, rewritten, shared for improvement, set aside and edited again with a fresh mind. — tap to centre the map on it A great memo cannot be written in aday or two: it is written,rewritten, shared for improvement,set aside and edited again with afresh mind. Last stated 9 years ago by 18 Apr 2018 JB Jeff Bezos — holds since 2018-04-18 — tap for who they are Same subject: Meetings should start with everyone silently reading a narratively structured six-page memo. — tap to centre the map on it Meetings should start with everyonesilently reading a narrativelystructured six-page memo. Last stated 9 years ago by 18 Apr 2018 JB Jeff Bezos — holds since 2018-04-18 — tap for who they are Same subject: When a slight inconsistency makes something significantly clearer, choose clarity — tap to centre the map on it When a slight inconsistency makessomething significantly clearer,choose clarity SK Steve Krug — holds since c. 2000–2014 — tap for who they are
same subject or similar wordinga cloud: claims about one subject, named for itbar: when it was last stated, on a scale from 2007 to today (stretched back to the oldest claim here) — full is todaya face: someone who holds the claim — tap it for who they are

At the centre Document similarity scores across a corpus tend to cluster bimodally near identical and near unrelated, rather than spreading evenly. Last stated 3 Jul 2024 · 2 years ago Holds Nelson Elhage Read this korrent →