Hacker News: treebrained

New comment by treebrained in "GLM 5.2 beats Claude in our benchmarks"

treebrained — Mon, 29 Jun 2026 13:02:23 +0000

For math, even the frontier has shortcomings, and there is a steep drop from GPT 5.5 xhigh to anything else. The time wasted by less-than-SotA just isn't worth it.

iOS 26.5 update adds age verification in the UK

treebrained — Wed, 25 Mar 2026 22:05:17 +0000

Article URL: https://www.theverge.com/tech/900284/apple-ios-26-4-uk-age-verification

Comments URL: https://news.ycombinator.com/item?id=47523913

Points: 3

# Comments: 0

Russian Authorities Block Archive.today

treebrained — Tue, 24 Mar 2026 09:00:45 +0000

Article URL: https://techcrunch.com/2026/03/23/russian-authorities-block-paywall-removal-site-archive-today/

Comments URL: https://news.ycombinator.com/item?id=47500097

Points: 9

# Comments: 1

Procedures for Repair of Potholes in Asphalt-Surfaced Pavements

treebrained — Sun, 01 Feb 2026 07:08:03 +0000

Article URL: https://highways.dot.gov/media/7941

Comments URL: https://news.ycombinator.com/item?id=46844195

Points: 50

# Comments: 65

Digital Freedom Across the World

treebrained — Thu, 29 Jan 2026 10:31:55 +0000

Article URL: https://eylenburg.github.io/countries.htm

Comments URL: https://news.ycombinator.com/item?id=46808229

Points: 3

# Comments: 0

UK universities offered to monitor students' social media

treebrained — Thu, 09 Oct 2025 06:32:29 +0000

Article URL: https://www.theguardian.com/education/2025/oct/08/uk-universities-offered-to-monitor-student-social-media-for-arms-firms-emails-show

Comments URL: https://news.ycombinator.com/item?id=45524239

Points: 4

# Comments: 0

Fictitious Telephone Number

treebrained — Sat, 13 Sep 2025 09:43:56 +0000

Article URL: https://en.wikipedia.org/wiki/Fictitious_telephone_number

Comments URL: https://news.ycombinator.com/item?id=45230758

Points: 1

# Comments: 0

Subliminal learning: Models transmit behaviors via hidden signals in data

treebrained — Tue, 22 Jul 2025 18:02:11 +0000

Article URL: https://alignment.anthropic.com/2025/subliminal-learning/

Comments URL: https://news.ycombinator.com/item?id=44650840

Points: 208

# Comments: 40

Elpe, the Universal Package Installer

treebrained — Mon, 09 Jun 2025 07:48:38 +0000

Article URL: https://nest.pijul.com/pmeunier/elpe

Comments URL: https://news.ycombinator.com/item?id=44222281

Points: 1

# Comments: 0

AI for the Working Mathematician

treebrained — Sat, 22 Mar 2025 13:05:26 +0000

Article URL: https://connorgs.net/mundane_utility_maths/

Comments URL: https://news.ycombinator.com/item?id=43445445

Points: 6

# Comments: 0

Heliograph

treebrained — Sun, 10 Mar 2024 14:37:53 +0000

Article URL: https://en.wikipedia.org/wiki/Heliograph

Comments URL: https://news.ycombinator.com/item?id=39659467

Points: 2

# Comments: 0

NIST's Wall of Many Stones

treebrained — Sun, 03 Mar 2024 02:02:08 +0000

Article URL: https://www.nist.gov/blogs/taking-measure/test-time-nists-wall-many-stones

Comments URL: https://news.ycombinator.com/item?id=39577737

Points: 3

# Comments: 0

New comment by treebrained in "Thousands of AI Authors on the Future of AI"

treebrained — Tue, 09 Jan 2024 02:32:26 +0000

Thank you for this comment. It is great to hear an inside take.

Idle curiosity, but what NLP tools evaluate translation quality better than a person? I was under the (perhaps mistaken) impression that NLP tools would be designed to approximate human intuition on this.

Their results are also highly biased. Most senior researchers aren't going to waste their time filling this out (90% of people did not fill it out). They almost certainly got very junior people and those with an axe to grind. Many of the respondents also have a conflict of interest, they run AI startups.

The survey does address the points above a bit. Per Section 5.2.2 and Appendix D, the survey had a response rate of 15% overall and of ~10% among people with over 1000 citations. Respondents who had given "when HLMI [more or less AGI] will be developed" or "impacts of smarter-than-human machines" a "great deal" of thought prior to the survey were 7.6% and 10.3%, respectively. Appendix D indicates that they saw no large differences between industry and academic respondents besides response rate, which was much lower for people in industry.

Thousands of AI Authors on the Future of AI

treebrained — Mon, 08 Jan 2024 21:23:59 +0000

Article URL: https://arxiv.org/abs/2401.02843

Comments URL: https://news.ycombinator.com/item?id=38918366

Points: 85

# Comments: 112

Macaulay2: A software system for algebraic geometry

treebrained — Thu, 02 Nov 2023 13:49:06 +0000

Article URL: https://www.macaulay2.com/

Comments URL: https://news.ycombinator.com/item?id=38113473

Points: 1

# Comments: 0