HNHacker News
TopNewBestAskShowJobs

letmevoteplease

1,098 karma · joined August 14, 2018

submissionscomments
letmevoteplease··on Figure 03, our 3rd generation humanoid robot
I'd love to be greeted by a robot when arriving at a hotel. Of course there's the novelty factor, but even without that, self-checkouts show that many people prefer interacting with a machine over a human for service.

More importantly, who wants to stand behind a desk 8 hours a day and handle fussy customers? Probably some people, but the main motivation for the average hotel clerk is receiving money. Can we reorganize the economy so robots perform this kind of mundane work, while humans still receive money but can spend their time on more meaningful activities than standing behind a desk? I think a future like that is possible although it remains to be seen whether we will get it.

letmevoteplease··on A human metaphor for evaluating AI capability
The tweet is not an accurate summary of the original post. The person who said they talked to the organizer did not say that. And now we are relying on a tweet from a person who said they talked to a person who said they talked to an organizer. Quite a game of telephone, and yet you're presenting it as some established truth.
letmevoteplease··on A human metaphor for evaluating AI capability
The source of this claim is a tweet.[1] The tweet screencaps a mathematician who says they talked to an IMO board member who told them "it was the general sense of the Jury and Coordinators that it's rude and inappropriate for AI developers to make announcements about their IMO performances too close to the IMO." This has now morphed into "OpenAI deliberately ignored the requests of IMO organizers to not publish AI results for some time."

[1] https://x.com/Mihonarium/status/1946880931723194389

letmevoteplease··on Gemma 3 Technical Report [pdf]
I don't agree that textual, fictional explicit content involving minors is "fairly universally considered harmful". Such content is allowed on large platforms like Archive of Our Own or Japan's Shosetsuka ni Naro. I think "don't think it's harmful, but not willing to defend" is a pretty typical attitude.
letmevoteplease··on OpenAI O3-Mini
The article says "expert testers."

"Evaluations by expert testers showed that o3-mini produces more accurate and clearer answers, with stronger reasoning abilities, than OpenAI o1-mini. Testers preferred o3-mini's responses to o1-mini 56% of the time and observed a 39% reduction in major errors on difficult real-world questions. W"

letmevoteplease··on Trying out QvQ – Qwen's new visual reasoning model
ChatGPT, Claude and Gemini get everything correct except who is holding the bucket. Even the QvQ attempt in your screenshot would have seemed like complete magic a couple years ago.
letmevoteplease··on Llama.cpp Now Supports Qwen2-VL (Vision Language Model)
The o1 release blog post contains 8 full examples of o1 chains of thought (not the summarized versions visible to users). They're English.

https://openai.com/index/learning-to-reason-with-llms/#chain...

I have seen the summaries dip into completely random languages like Thai, so it might switch between languages occasionally.

letmevoteplease··on Software developer arrested in connection with murder of healthcare executive
The video (https://www.youtube.com/watch?v=uCvKPBebNTk) was taken down 33 minutes before it was set to premiere, and now the account (https://www.youtube.com/@PepMangione) is deleted. The thumbnail had "The Truth will set me free" written in binary. Screenshot: https://i.imgur.com/ovODSvx.png

Here is a mirror of the short video linked in the OP (no real content, just a countdown and date): https://files.catbox.moe/jxtf97.mkv

letmevoteplease··on Software developer arrested in connection with murder of healthcare executive
From the font and awkwarding positioning I'm guessing the date was overlayed on the video with a script or something (i.e. "two days after whenever this deadman's switch gets triggered")
letmevoteplease··on Sora is here
Shane Carruth (Primer) released interesting scripts for "A Topiary" and "The Modern Ocean" which now have no hope of being filmed. I hope AI can bring them to life someday. If we get tools like ControlNet for video, maybe Carruth could even "direct" them himself.
letmevoteplease··on Bird flu in Canada may have mutated to become more transmissible to humans
This is pretty clearly a joke: "Really, I don't really wash my hands ever. Germs are not a real thing. I can't see them, therefore they not real. I can't get sick."

https://x.com/RonFilipkowski/status/1856494554624442433

letmevoteplease··on Vivek Ramaswamy on X: "Will entire agencies be deleted? Answer: yes
The Republican Party Platform says: "We are going to close the Department of Education in Washington, D.C. and send it back to the States, where it belongs, and let the States run our educational system as it should be run."

https://www.presidency.ucsb.edu/documents/2024-republican-pa...

Trump also mentioned it frequently in his speeches: " 'I say it all the time, I’m dying to get back to do this. We will ultimately eliminate the federal Department of Education,' he said in September during a rally in Wisconsin. "

https://www.cnn.com/2024/09/20/politics/department-of-educat...

letmevoteplease··on Boston Dynamics robot Atlas goes hands on [video]
Tesla has footage of Optimus working autonomously. Not bad progress for something that has been under development for two years.

https://x.com/Tesla_Optimus/status/1846797392521167223

letmevoteplease··on Understanding the Limitations of Mathematical Reasoning in LLMs
I prompted o1 with "analyze this problem word-by-word to ensure that you fully understand it. Make no assumptions." and it solved the "riddle" correctly.

https://chatgpt.com/share/6709473b-b22c-8012-a30d-42c8482cc6...

letmevoteplease··on Vision language models are blind
No reason to believe that. Open source VLMs can do OCR.[1]

[1] https://huggingface.co/spaces/opencompass/open_vlm_leaderboa...

letmevoteplease··on KidPix
I can't figure out how to clear the canvas.
letmevoteplease··on Horseshoe crab counting with New York’s citizen scientists
Pictures of the harvesting process are... interesting.

https://i.imgur.com/IKxgi1U.png

letmevoteplease··on WinDirStat – Windows Directory Statistics
Try Everything 1.5a - an "alpha" version with many improvements, in development for years but inexplicably hidden away on their website. Never experienced any instability.

https://www.voidtools.com/forum/viewtopic.php?t=9787

letmevoteplease··on China’s Moon atlas is the most detailed ever made
Scaled to 50% (19.2MB): https://files.catbox.moe/egqh3x.jpg
letmevoteplease··on How China Built BYD
According to this list, even its current state, the Nevada Gigafactory has the most floorspace of any warehouse in the world (5.3 million ft², followed by Boeing with 4.3 million ft²). Not entirely unimpressive.

https://www.avantauk.com/top-14-largest-warehouses-in-the-wo...

letmevoteplease··on The Empire Strikes Back First Draft by Leigh Brackett (1978) [pdf]
Here is an article about this case:

https://www.battlegrip.com/star-wars-writer-stabs-best-frien...

letmevoteplease··on Japan's first-ever soft lunar landing with SLIM spacecraft [video]
The official JAXA streams are here:

Japanese: https://www.youtube.com/watch?v=Udh6kvjZYC8

English: https://www.youtube.com/watch?v=nvXLt3ET9mE

Edit: It was supposed to touch down at this point, but they are pausing for 30 minutes to check the status of the craft. Doesn't look hopeful to me.

Edit 2: They now say it may take up to 2 hours to confirm the situation.

Edit 3: The craft landed successfully and is communicating properly, but the solar cell is not generating electricity, so it is being operated on low battery power. Lunar Excursion Vehicle 1 and 2 have been successfully separated.

letmevoteplease··on Hidden Changes in GPT-4, Uncovered
If companies are required to purchase licenses for everything they train on, it will guarantee that only huge corporations with deep pockets can produce powerful models. Microsoft will be slightly inconvenienced, Stability AI will be destroyed. Some artists might get a payday, but most of the money will go to companies with large copyright libraries like Getty. The general quality of all models will decrease. I don't see any other possible outcome.
letmevoteplease··on Warning from OpenAI leaders helped trigger Sam Altman's ouster
I stayed up late to follow the drama and I don't even work there.
letmevoteplease··on Why does sleep become more elusive as we age?
Also alarm clocks, caffeine, unnatural sleep schedules. This site[1] has lots of good information, with this basic (unfortunately impossible for many) recommendation:

"There is only one formula for healthy and refreshing sleep: Go to sleep only when you are very tired. Not earlier. Not later. Wake up naturally without an alarm clock."

[1] https://supermemo.guru/wiki/Science_of_sleep

letmevoteplease··on $500 a day to pretend to be a model: The big business behind OnlyFans 'chatters'
I wouldn't say that sex work is fundamentally immoral, but using deception to manipulate people into giving you money is hard to defend. Of course this occurs in many other industries, too.
letmevoteplease··on Tesla sues Swedish state agency over number plate blockage
If a company is legally required to use PostNord and PostNord can legally refuse to deliver mail, it effectively gives PostNord the power to control any company that wants to continue receiving mail. Regardless of whether they are using that power for good or evil here, I don't think it's a power they should have.
letmevoteplease··on Bard Extensions
Here is GPT-4 for reference. I can't judge the quality. https://chat.openai.com/share/96d4f976-627f-4cc3-8716-3557d5...
letmevoteplease··on Bard Extensions
Bard has had access to Google caches since launch. I asked it about my 8-follower Twitter account and it quoted my first Tweet and the registration month correctly. It also hallucinated my gender and described me as "a popular figure in the French-speaking Twitter community" (because my name is a French word, I guess).
letmevoteplease··on Before Altman’s ouster, OpenAI’s board was divided and feuding
According to the paper, Anthropic's superior safety approach was deliberately delaying release in order to avoid “advanc[ing] the rate of AI capabilities progress." OpenAI is criticized for kicking off a race for AI progress by releasing ChatGPT to the public.

[1] https://cset.georgetown.edu/wp-content/uploads/CSET-Decoding...

← PreviousPage 3 of 6Next →