intended's favorites | Hacker News

1.		My trick for getting consistent classification from LLMs (verdik.substack.com)
		318 points by frenchmajesty 15 days ago \| 70 comments
2.		Rupert's snub cube and other Math Holes (tom7.org)
		148 points by QuadmasterXLII 42 days ago \| 10 comments
3.		This map is not upside down (maps.com)
		365 points by aagha 40 days ago \| 514 comments
4.		Defeating Nondeterminism in LLM Inference (thinkingmachines.ai)
		345 points by jxmorris12 48 days ago \| 130 comments
5.		Almost anything you give sustained attention to will begin to loop on itself (henrikkarlsson.xyz)
		773 points by jger15 54 days ago \| 223 comments
6.		Notes on Managing ADHD (borretti.me)
		633 points by amrrs 58 days ago \| 330 comments
7.		Context Rot: How increasing input tokens impacts LLM performance (trychroma.com)
		260 points by kellyhongsn 3 months ago \| 59 comments
8.		Writing a good design document (grantslatton.com)
		581 points by kiyanwang 85 days ago \| 147 comments
9.		Nobody knows how to build with AI yet (worksonmymachine.substack.com)
		526 points by Stwerner 3 months ago \| 407 comments
10.		Engineered Addictions (masonyarbrough.substack.com)
		720 points by echollama 4 months ago \| 451 comments
11.		Accumulation of cognitive debt when using an AI assistant for essay writing task (arxiv.org)
		391 points by stephen_g 4 months ago \| 426 comments
12.		The Illusion of Thinking: Strengths and limitations of reasoning models [pdf] (cdn-apple.com)
		488 points by amrrs 4 months ago \| 270 comments
13.		Why Good Ideas Die Quietly and Bad Ideas Go Viral (newyorker.com)
		70 points by samizdis 5 months ago \| 120 comments
14.		Trying to teach in the age of the AI homework machine (solarshades.club)
		481 points by notarobot123 5 months ago \| 649 comments
15.		What makes a good engineer also makes a good engineering organization (2024) (moxie.org)
		337 points by kiyanwang 5 months ago \| 139 comments
16.		People are losing loved ones to AI-fueled spiritual fantasies (rollingstone.com)
		182 points by wzm 5 months ago \| 161 comments
17.		I decided to pay off a school’s lunch debt (huffpost.com)
		589 points by dredmorbius 5 months ago \| 867 comments
18.		I wrote a book called “Crap Towns”. It seemed funny at the time (samj.substack.com)
		395 points by url 6 months ago \| 314 comments
19.		AI assisted search-based research works now (simonwillison.net)
		283 points by simonw 6 months ago \| 147 comments
20.		[flagged] We're Raising Kids to Prefer AI over People–and No One's Noticing (substack.com)
		46 points by softchaos 6 months ago \| 79 comments
21.		A Reddit bot drove me insane (posthuman.blog)
		699 points by erhmmmm 6 months ago \| 364 comments
22.		Ask HN: Do your eyes bug you even though your prescription is "correct"?
		340 points by jbornhorst 7 months ago \| 326 comments
23.		Introduction to Stochastic Calculus (jiha-kim.github.io)
		444 points by ibobev 8 months ago \| 70 comments
24.		Why is it so hard to buy things that work well? (2022) (danluu.com)
		693 points by janandonly 10 months ago \| 577 comments
25.		What can LLMs never do? (strangeloopcanon.com)
		460 points by henrik_w on April 27, 2024 \| 374 comments
26.		Non-determinism in GPT-4 is caused by Sparse MoE (152334h.github.io)
		397 points by 152334H on Aug 4, 2023 \| 181 comments
27.		Patterns for building LLM-based systems and products (eugeneyan.com)
		436 points by 7d7n on Aug 2, 2023 \| 55 comments
28.		Bad numbers in the “gzip beats BERT” paper? (kenschutte.com)
		381 points by ks2048 on July 17, 2023 \| 130 comments
29.		Gzip beats BERT? Part 2: dataset issues, improved speed, and results (kenschutte.com)
		248 points by JoeyBananas on July 29, 2023 \| 51 comments
30.		Nuanced communication usually doesn't work at scale (twitter.com/danluu)
		725 points by tagolli on Jan 29, 2022 \| 271 comments
		More