#Accuracy in AP

[technicalfika]

Mastering Accounts Payable: 15 Common Interview Questions and Expert Answers

Preparing for an Accounts Payable (AP) interview is a crucial step in securing a position in this vital finance function. To help you excel in your interview, we’ve compiled a comprehensive guide that covers the 20 most common AP interview questions, each paired with expert answers. In this article, you’ll gain valuable insights into the world of AP and enhance your chances of acing your…

[ssruis]

Forgot about this yesterday first try let’s go woohoo

[springeneticist]

accidentally made myself angry thinking about gacha rhythm scoring dynamics. cause why can i only name one mobile idol rhythm game where it gives you a technical score separately to your gacha slop score after playing a show

[not-poignant]

I was just wondering, in the UTB universe, how are peak alphas differentiated from regular alphas? You say Caleb acts like a peak alpha, and Augus is mild as a peak alpha. So how are peak alphas recognized as such?

Hi hi!

So all secondary sexes (omega / beta / alpha / peak alpha) are determined by a combination of blood + genetics + physiological markers (i.e. normative penis size / presence of certain glands / knotting etc) + active hormone levels - and for peak alphas: a range of unique abilities they have that no other secondary sex possess (like the ability to use alpha persuasion on other alphas, and the ability to use an innate persuasion/hypnosis ability on everyone).

This is why they test Efnisien to make sure he's an omega because of what he says in the beginning of the story - there's a biological component to secondary sex (and we see this too in the story when Temsen talks about normal hormone level ranges in the chapter where he talks about Efnisien's hormone levels).

The genetic / biological components of secondary sex are talked about in the story quite a bit (Efnisien being omega according to both blood tests but also genetically is mentioned around 10+ times).

So Efnisien will always be biologically an omega despite his active hormone levels (and they do have to take that into account when they supplement his hormones/ardolphogen levels, even as they recognise him as an alpha as well - which is about identity and not biology). And Caleb is biologically be an alpha despite acting more like a peak alpha. And Augus is biologically a peak alpha, despite his behaviours. Acting like a peak alpha isn't the same as being one!

As to Efnisien being treated like an alpha, that's partly because he has been scientifically experimented on, mutilated, had organs removed, and changed so much that he just is no longer really exactly anything at all completely. He can no longer make larentin, he no longer has Kaeper glands or lare glands, he developed unique alpha abilities because he was forced onto ardolphogen years before even alphas start producing it in large quantities, etc. And partly because he's been raised an alpha. And partly because he sees himself as one.

Caleb doesn't see himself as a peak alpha, even though he can behave like one. Augus doesn't see himself as a softer alpha, even though he behaves like one.

The answer is both 'with science' and 'it's complicated' depending on the person :D

[littleaipom]

new fursona! (not to replace my usual one) It is time that i embrace the charm of monkey :D

I'm naming her Zivi

[xartus]

Here's some reconstructions ACTUALLY based on historical findings of Viking Age female divinatory ritual specialists!!!!! See how they look NOT like the previous post!!!!

[majorshatterandhare]

I started a piece of art today which is based on some plant species* that I think would be good to colonize Tim in @gunpowder-tim’s headcanon of the Persephone Tim headcanon; so it’s art based on a headcanon of a headcanon of a headcanon 😅 [sweat simle emoji].

It’s gonna take a long time I think, but I am planning on posting it here even though it’s just gonna be plants and no Tim (because I am much better at drawing these little doodle plants than drawing people).

*so the art is basically of real species we have on Earth, but I maintain that they wouldn’t have the same plants on the City, so in my brain his plants are just similar to these ones.

[slocumjoe]

It's so funny to think how we both have the incredible specific scenario of 'my sole survivor was stuck somewhere while serving in the military and had to eat his fellow soldiers and now he's a Cannibal'. Its like that Doofinshmirtz meme. Its weird that this happened twice. Here I was thinking I had galaxy brained when I did this a month ago and bam!! You did it too!!

8 END is both a blessing and in the right wrong circumstances, a curse

[apotelesmaa]

I am merging my consciousness with hatsune miku as we speak.

[kyokosskyblog]

FIRST TRY WITH NO ACCURACY BOOST

[link-lonk]

PAIN AND SUFFERING NO MORE! I DID IT AND IT'S ON VIDEO!!!

This is the first expert song I've gotten an all perfect on!!

[leonstoenailunderhisbed]

I don’t think people understand how smart Leon actually is. That man had high marks from the police academy, hence why Chief Irons (in the orientation letter Leon has in RE2R) says that his grades are commendable and that R.P.D. are proud to have him on the force.

This got me thinking a lot because at first I thought being a police officer didn’t take much given that anyone today can be in the force.

But that wasn’t the case in 1998.

I did a little research because I thought it would be interesting to see just what Leon had to go through to become a police officer.

Back in the 90s, cellphones and modern technology didn’t exist such as DNA identification and body cams/car cams. This definitely made the job a bit harder than it is today because there was a need for more evidence to be collected and the overuse of your brain. Nowadays, technology is an important factor in the police force and almost everything is done by the computer now.

So let’s picture this: It’s the late 90s right before he got sent to Raccoon City. He’s in the academy and he has to go through training. Especially with weapons since most academies switched from revolvers to semi-auto guns (already something a bit modern for that time).

For those who’ve played the game, when you go into the Shooting Range room, you can clearly see just how old the room is compared to modern shooting ranges. Not only is the design of the target paper old, it’s also very simplistic compared to today’s (in 2024, most markings have numbers and more lines for accuracy than back in the day).

This meant that Leon had to train a good amount of time to perfect his aim. It also meant that he had to go through driving training—which was mostly Emergency Vehicle Operations Courses (EVOC: safe and defensive driving for cops in other words)

I’d like to think that his determination (when he told Ada that the reason he joined the force was for people like Emma and Gunshop owner) really helped him advance through his academic route of the training and I’d like to believe that he go high scores because of that.

The 90s were a pivotal time for new policies to be introduced in police academies. When Leon was a kid, presumably during the 80s, he probably saw just how different it was back then than it is now (in 1998) lots of “new” technology were introduced to him when he first started the training. And he probably had to adapt quickly to the technologies and new techniques.

Leon is quick on his feet, he grasps a lot of things and I’m tired of people making him out to be as some dumb blonde with muscles. He’s very smart and we see that throughout the games and films.

I wouldn’t be surprised if he was GT when he grew up.

EDIT: MB YALL😭 GT is a program for students K-12 where they’re put in advanced classes like AP or IB. It stands for Gifted and Talented (something like the Magnet Program in some schools in the US)

[secretarysong]

Mayzuketober Day 6: Masquerade

oh she is gonna SPIN his ass like a WINDMILL

mask details + rambling under the cut ..i don't want the cropping to be Teeww obnoxious

can you guys tell i based them off of a phoenix and a peacock ... i don't make super ornate/detailed stuff often so this was a nice change of pace teehee. also do NNOOOOTTTT QUUOTE ME on the historical accuracy of it *starts scratching my ass like an ape* I'm Just here to have fun man. I think after the month is over i will draw the full outfits i envisioned for this prompt . Okaaayyyy thank you for reading byebye

[foone]

Alternative names for humanity along the lines of "Homo sapiens" (Wise man) and "Pan narrans" (Storytelling Chimpanzee) that I'm too lazy to look up/make up Latin for:

chef ape

throwing ape

walking ape

The idea being that we're apparently unique in the animal kingdom in that we cook our food, so we're the Chef Apes. We're also one of the best animals at throwing things: humans have more accuracy and strength when throwing stuff than other apes, by a long shot

And apparently our ability to walk slowly for ages was key to our early survival as persistence predators. We can't outrun a gazelle or mammoth or whatever, but we don't tire easily and so we can just keep following it until it runs out of stamina

Pan basipila: the baseball playing Bonobo

If only baseball had a cooking element, it would be the perfect Human Sport.

We need to devise a sport where you cook something, follow someone for a long time, and then throw it at them.

The most human thing is the surprise pie to the face

Also as much as I like Terry Pratchett's suggestion of "Pan narrans" I wouldn't be surprised if we turn out to not be the only animal that tells stories...

Elephants. I bet elephants do.

Like, there was that case where an injured elephant went to a ranger station for help. One it had never been to before, but other elephants had.

The theory being then that some other elephant had told this elephant "hey if you're hurt, go here, the humans will help"

That, combined with how they have burial rituals (some which might indicate there's an elephant religion!), and that we're working on figuring out how elephants communicate...

It wouldn't surprise me if we learn sometimes in the next decade or two that "oh yeah, elephants tell stories too. They've got FICTION."

So "Pan narrans" isn't what I'd want to bet on as our uniquely human thing.

But at the end of the day, maybe the whole idea of there being a uniquely human thing is, in itself, just another story we're telling.

So maybe it is a good fit after all.

But I especially like the idea that we're the Baseball Ape because I have this image in my head of a galactic council of aliens. Some angry alien who looks like Cthulhu had a baby with a spider has the floor, and they're ranting about "why do the Hu-mons deserve a seat?"

The Crogath are stronger, the Eldru are smarter, the Cybernetic Essense lives longer, the Dromans go farther and faster, the Moltriri have us beat in fiction and poetry, what is so special about these damn bipedal fleshbags that makes them unique in the universe?

And then WHAM. Right between the eyes. A handheld translator device, a bit bigger than a modern smartphone, beans the speaker out of nowhere.

And there's an (untranslated) yell in the chamber as the prime representative calls for order.

"WE CAN THROW, MOTHERFUCKER!"

(it takes a while to properly explain the insult. Crogathi (especially drones) don't really have mothers or sexual reproduction, so they don't really get why that would be an insult. It's finally translated as something like "bud-biter")

and it's true. even after the World Series becomes the Galactic Series, no non-human team ever manages to win.

The Eldrul Librarians almost make the cut in 2486 but accidentally piss off the ghost of Colonel Sanders and end up inheriting the Hanshin Tigers' curse.

alien textbooks describe The Colonel as some kind of human patron deity of baseball and cooked avian food, who should not be disrespected at all costs, or his vengeance from his place beyond the grave will be swift and punishing

(they're right)

"Look, we can't PROVE he was why Gemini Noctis went supernova unexpectedly, but given the protests that had happened right beforehand, and the incredible powers ascribed to the human spirits, do you really want to risk it?"

the funniest possible future: humanity gets a key place in galactic politics because we're never able to adequately convince the universe at large that our ghost stories are just that, stories, and they're terrified shitless that we'll unleash spectral torment on them

"humans? look man, living humans are a pushover. you can easily rip them in half, crack their planets with a quark bomb, their ships are little more than tin cans with a tachyon drive taped on the side. but it's not the living humans you have to worry about... it's the ghosts."

"humans are a bit like the Nontilek, with a two-stage lifespan, a grub and an adult. What you think of as "adult" humans is just their infant stage, and they only fully transform once they "die". Once fully hatched into Ghost form, their powers are almost limitless."

you want humans off a colony planet and bomb them from orbit? good luck, now you have a few million ascended humans who can pass through solid matter and can't be killed, and they will never rest until you and your descendants are gone or dead.

you don't believe me? look at this: One of their most popular stories is about them building an empire that spanned a large chunk of their little planet, then having it MURDER THEIR OWN GOD.

It only worked for a few revolutions, and he just came back, promising that one day all of them would join him in the next phase of their lifespan.

They still, to this day, thousands of orbits later, erect little statues of the means they used to execute their deity.

not even the Crogathi, who literally worship death itself, tell stories that frightening to their newly hatched grubs.

Humans are scary, man, stay away and just give them whatever they want.

the rest of the alien's education on the dangers of humans is just a selection of human movies. the sixth sense, poltergeist, ghostbusters, the shining, the devil's backbone, and, of course, field of dreams.

ghosts AND baseball? it's everything they're scared about humans all in one package!

the obvious twist you could do, of course, is simple:

the aliens are right.

humans are a two-phase species where the elder form has immense power but leaves communication and decision making to the younger form, which will be confused and angry if you acknowledge the presence of their elder-stage members among them.

this often leads to them cutting off contact or their elder-stage members causing immense damage through seeming "accidents" on the contacting vessel. This is believed to be some kind of religious prohibition that they are not able to explain.

so it's official contact protocol to pretend you cannot perceive the elder-stage humans among them, and to give them what they want to avoid possible retribution.

No means to combat elder-stage humans has yet been found, and the limits of their power is not known.

All alien captains are required to study the fate of the SS Ennolon, which contacted a lone human craft in the galactic year of 12,783. They had initiated contact and were getting along fine, until the human showed the Droman captain a picture of their "late father".

Captain Droless, accounting for the difficulty in telling humans apart, then pointed at the father sitting in a chair nearby and said "That is them, correct?".

The human looked at the chair, reacted in confusion, then anger, and asked the contacting crew to immediately leave.

It was another 400 cycles before contact could be reestablished between the Droman Federation and the Human Alliance.

the intergalactic guide describes humans as a powerful race of immortal energy beings who have the strange habit of sending their larvae out on missions around the galaxy, occasionally contacting other races, but refusing to acknowledge their elders, except in stories

they seem to frequently put their young in dangerous situations without lifting a hand to help, so this is suspected to be some sort of pilgrimage or coming-of-age ritual.

(From a twitter thread on October 1st, 2022)

[littybeech]

Ten things the Twilight animated series can do better than the original movies (if they care enough to, please add all your own points to this as I will, too.) :

1. Include ALL of the book scenes, without having to change them to make them easier to film (for obvious reasons, since it’ll be animated they shouldn’t have much issue there.)

2. Their daughter, Remoulade can actually look like the book described her (or better or worse) and not the CGI monstrosity we got twelve years ago.

3. Make all the characters fit their book looks to complete accuracy. Alice was 4’10 and the movies made her 5’5. They let Edward have brown hair when he’s a ginger. Jacob and the wolf pack had better be above 6’5! I want 23 year old Carlisle and 26 year old Esme playing parents to like three legal adults and two 17 year olds.

4. The parentification of Bella Swan, by both of her parents, mustn’t be ignored. Bella was making sure the bills were getting paid by the time she was 10 cause her mom was ‘too flighty and distracted’ to do things like that consistently, she also learned to cook and clean early on too because her mom’s cooking was inedible and she’d improperly mix cleaning solutions dangerously. She got a job at 14 and took care of the groceries and any other issue necessary on top of being in AP classes in school. By the time she moved in with Charlie, she was basically more of a parent than he was. He did the bare minimum to ensure her car’s safety by installing snow chains on her tires without telling her and she cried because she wasn’t used to being taken care of.

5. How she cries when she’s angry is peak girlhood and I hate that they got rid of that in the movies. She stomped her feet bro, like c’mon she was so angry and anxious and annoyed all the time and they only focused on her angst.

6. Integrate aspects of Midnight Sun too, maybe incorporate them both so we can get a more full story. Twilight on its own was a bit of a snooze fest compared to Midnight Sun. We’re gonna pretend that Edward hunting and eating Esme’s abusive ex-husband isn’t hot af?

7. Add in more character’s back stories. Alice’s abusive father and stepmother, Emmett’s gambling and womanizing as a human in Tennessee, etc.

8. Unfortunately we need to make it clear that Charlie’s not winning any father of the year awards here either. He congratulated Jacob after forcing a kiss on her and joked about Jacob pressing assault charges on her after she breaks her hand punching him away.

9. Maybe ask why the Cullens…’need’ so desperately to come back to Forks every couple of years? Like they’re multi-billionaires, they’re immortal and there are plenty of gloomy, cloudy places they could live in semi-permanently. Why Forks when they know the Quileutte tribe knows what they are and they know it makes them so uncomfortable? It just seems unnecessary.

10. More of Edward reading Mike Newton’s inner-monologue. I know it would depend on whose perspective we get but I think the idea is hilarious.

[mariacallous]

On Saturday, an Associated Press investigation revealed that OpenAI's Whisper transcription tool creates fabricated text in medical and business settings despite warnings against such use. The AP interviewed more than 12 software engineers, developers, and researchers who found the model regularly invents text that speakers never said, a phenomenon often called a “confabulation” or “hallucination” in the AI field.

Upon its release in 2022, OpenAI claimed that Whisper approached “human level robustness” in audio transcription accuracy. However, a University of Michigan researcher told the AP that Whisper created false text in 80 percent of public meeting transcripts examined. Another developer, unnamed in the AP report, claimed to have found invented content in almost all of his 26,000 test transcriptions.

The fabrications pose particular risks in health care settings. Despite OpenAI’s warnings against using Whisper for “high-risk domains,” over 30,000 medical workers now use Whisper-based tools to transcribe patient visits, according to the AP report. The Mankato Clinic in Minnesota and Children’s Hospital Los Angeles are among 40 health systems using a Whisper-powered AI copilot service from medical tech company Nabla that is fine-tuned on medical terminology.

Nabla acknowledges that Whisper can confabulate, but it also reportedly erases original audio recordings “for data safety reasons.” This could cause additional issues, since doctors cannot verify accuracy against the source material. And deaf patients may be highly impacted by mistaken transcripts since they would have no way to know if medical transcript audio is accurate or not.

The potential problems with Whisper extend beyond health care. Researchers from Cornell University and the University of Virginia studied thousands of audio samples and found Whisper adding nonexistent violent content and racial commentary to neutral speech. They found that 1 percent of samples included “entire hallucinated phrases or sentences which did not exist in any form in the underlying audio” and that 38 percent of those included “explicit harms such as perpetuating violence, making up inaccurate associations, or implying false authority.”

In one case from the study cited by AP, when a speaker described “two other girls and one lady,” Whisper added fictional text specifying that they “were Black.” In another, the audio said, “He, the boy, was going to, I’m not sure exactly, take the umbrella.” Whisper transcribed it to, “He took a big piece of a cross, a teeny, small piece … I’m sure he didn’t have a terror knife so he killed a number of people.”

An OpenAI spokesperson told the AP that the company appreciates the researchers’ findings and that it actively studies how to reduce fabrications and incorporates feedback in updates to the model.

Why Whisper Confabulates

The key to Whisper’s unsuitability in high-risk domains comes from its propensity to sometimes confabulate, or plausibly make up, inaccurate outputs. The AP report says, "Researchers aren’t certain why Whisper and similar tools hallucinate," but that isn't true. We know exactly why Transformer-based AI models like Whisper behave this way.

Whisper is based on technology that is designed to predict the next most likely token (chunk of data) that should appear after a sequence of tokens provided by a user. In the case of ChatGPT, the input tokens come in the form of a text prompt. In the case of Whisper, the input is tokenized audio data.

The transcription output from Whisper is a prediction of what is most likely, not what is most accurate. Accuracy in Transformer-based outputs is typically proportional to the presence of relevant accurate data in the training dataset, but it is never guaranteed. If there is ever a case where there isn't enough contextual information in its neural network for Whisper to make an accurate prediction about how to transcribe a particular segment of audio, the model will fall back on what it “knows” about the relationships between sounds and words it has learned from its training data.

According to OpenAI in 2022, Whisper learned those statistical relationships from “680,000 hours of multilingual and multitask supervised data collected from the web.” But we now know a little more about the source. Given Whisper's well-known tendency to produce certain outputs like "thank you for watching," "like and subscribe," or "drop a comment in the section below" when provided silent or garbled inputs, it's likely that OpenAI trained Whisper on thousands of hours of captioned audio scraped from YouTube videos. (The researchers needed audio paired with existing captions to train the model.)

There's also a phenomenon called “overfitting” in AI models where information (in this case, text found in audio transcriptions) encountered more frequently in the training data is more likely to be reproduced in an output. In cases where Whisper encounters poor-quality audio in medical notes, the AI model will produce what its neural network predicts is the most likely output, even if it is incorrect. And the most likely output for any given YouTube video, since so many people say it, is “thanks for watching.”

In other cases, Whisper seems to draw on the context of the conversation to fill in what should come next, which can lead to problems because its training data could include racist commentary or inaccurate medical information. For example, if many examples of training data featured speakers saying the phrase “crimes by Black criminals,” when Whisper encounters a “crimes by [garbled audio] criminals” audio sample, it will be more likely to fill in the transcription with “Black."

In the original Whisper model card, OpenAI researchers wrote about this very phenomenon: "Because the models are trained in a weakly supervised manner using large-scale noisy data, the predictions may include texts that are not actually spoken in the audio input (i.e. hallucination). We hypothesize that this happens because, given their general knowledge of language, the models combine trying to predict the next word in audio with trying to transcribe the audio itself."

So in that sense, Whisper "knows" something about the content of what is being said and keeps track of the context of the conversation, which can lead to issues like the one where Whisper identified two women as being Black even though that information was not contained in the original audio. Theoretically, this erroneous scenario could be reduced by using a second AI model trained to pick out areas of confusing audio where the Whisper model is likely to confabulate and flag the transcript in that location, so a human could manually check those instances for accuracy later.

Clearly, OpenAI's advice not to use Whisper in high-risk domains, such as critical medical records, was a good one. But health care companies are constantly driven by a need to decrease costs by using seemingly "good enough" AI tools—as we've seen with Epic Systems using GPT-4 for medical records and UnitedHealth using a flawed AI model for insurance decisions. It's entirely possible that people are already suffering negative outcomes due to AI mistakes, and fixing them will likely involve some sort of regulation and certification of AI tools used in the medical field.