ana. (@inconsistente) header
ana. (@inconsistente) profile picture

free & easy gentle gentle

CDMX
Joined June 2010
Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)Photo by ana. (@inconsistente)
So, what exactly is LLM writing? How is it different from how humans write? If you look at it closely, it's not just the groups of three, or "it's not X, it's Y"; it's the whole set of characteristics that usually go together. To study this, I used the same algorithm that Pangram uses to train its classifier. It's called "LLM twins." It works like this: I downloaded 200 samples of verifiably human writing across all genres, then asked a language model to create a short description of each, and then to fire up separate agents, which had to re-create the pieces from the descriptions alone and make them the same length as the originals. As a result, I obtained 200 pairs of pieces of writing, each pair sharing a genre, a length, and a rhetorical goal. For example, a human-written press release about a Boca Raton condominium hosting the filming of "Selling New York," and an LLM-written press release about the same event. Pangram generates millions of those. I only had 200, but that allowed me to measure the main differences between humans and LLMs. Here are the main differences: 1. LLMs rarely let two adjacent clauses or sentences share a noun. A human would write, "I stepped into a corridor. The corridor was dark and narrow." An LLM would say, "I stepped into a corridor. Dark. Narrow." 2. LLMs show much more diversity in vocabulary. A human would say "red" or "dark-red." An LLM would write "plum-coloured." 3. LLMs write fewer words per clause. A human would write, "We checked the reported numbers, our investigation found no discrepancies." An LLM would write, "We checked the numbers, no discrepancies found." 4. LLMs write shorter sentences. This is self-explanatory, I believe. 5. LLMs seem to avoid proper nouns (people, places) if they can. A human would write, "A distraught displaced woman, Hajiya Mairo Abdullahi, was at the weekend seen holding the picture of her seven-year-old child." An LLM would, among other things, omit the woman's name if it isn't needed for the narrative and doesn't recur later in the text. The LLM version would go something like: "A displaced woman, struggling with grief, was clutching a photograph of her seven-year-old child." Even if the person is named, an LLM is more likely to avoid using the name when attributing speech to them. A human would write, "Ms Abdullahi said," while an LLM would write, "The woman said." 6. LLMs also avoid repeated 3-word sequences, whereas humans are okay with them. A human would write "the Kenya cases"; an LLM would rephrase it as "cases from Kenya" or "the cases," or structure the sentence differently to avoid repetition. 7. LLMs operate in a narrower band of word frequencies: more often than humans, they avoid very common words (they are more likely to say "plum-coloured" instead of "red"), but they also avoid rare words (they are unlikely to say "incarnadine" instead of "red"). 8. LLMs use fewer numbers. This could be specific to news articles, but LLMs prefer to say "many," "several," or "thousands" when the exact number isn't essential beyond conveying the ballpark. A human would put an actual number. And finally, an aspect that captures the specific feeling of hollowness that encountering LLM writing brings. 9. Each new sentence in an LLM twin is more likely to introduce a new common noun ("displacement," "war," etc.). But it's less likely to introduce a new, specific, verifiable fact. A human would write: "Parents want to know if their children are alive and when they will come home. An inquiry has been launched by the Nigerian government..." An LLM would write: "For the parents, the question is simple. They want to know if their children are alive and when they are coming home." This is, of course, from studying 200 samples. Pangram has millions, and its neural networks, as unexplainable as this technology is, can probably be contextualised as checking tens of thousands of such "tics." Importantly, none of these is consequential on its own, but in LLM writing all these features tend to go together. That's why people who work a lot with LLMs can often (but not always) recognise their style.
9
23
4
331
15,708
Replying to @debrafig
No tiene letrero aún pero está en la esquina de Juan Álvarez y Enrique Díaz de León en Santa Tere
32
"No pasa nada".
5
294
Escenas del fin del mundo o la injusticia de creer que esto es, como dicen nuestras autoridades, “brindar apoyo”.
37
149
18
296
10,489
Acabo de ver que en un restaurante que apenas van a inaugural, metieron decenas de jaulas con palomas negras. ¿A alguien se le ocurre para qué pueden ser? Estoy muy impresionada.
1
4
640
En pocas palabras, va a haber más agua, menos tandeos, pero el agua volverá a salir contaminada, pues el sistema antiguo sigue sin ser saneado en su totalidad. ¿Usted prefiere quejarse porque no hay agua o porque el agua apesta y enferma? 🤷‍♀️
Con el encendido del primero de los tres sistemas de bombeo se retoman actividades en la planta de Las Pintas, lo que permitirá recuperar el volumen de agua y estabilizar el esquema de Rotación de Servicio que aplica en algunas colonias de la Zona Metropolitana de Guadalajara.
17
58
2
210
6,815
Replying to @Batistadriana
Me choca eso
1
10
Un subset de la progresía clasemediera se dice agnóstica pero opera con culpa católica, y en parte por eso hay un boom de escritura mediocre pero que habla de mundos (barrios) a los que sólo han visto de lejos. Es la finísima condescendencia de las bajas expectativas.
4
34
882
Me da flojera sobre todo el discurso del progreso. "Es que no puedes oponerte al progreso", dice mucha gente sobre la "IA". No sé si ya pensaron que quizá no compartimos la misma definición de "progreso"? Por qué es progreso tecnificar absolutamente todo? Zzz.
Me da tanto gusto ver a cada vez más gente criticando abiertamente a la "IA" y sus implicaciones. Hubo un momento en el que parecía que todos estaban fascinados pero va cayendo la máscara 🤖☠️
2
20
639
Me da tanto gusto ver a cada vez más gente criticando abiertamente a la "IA" y sus implicaciones. Hubo un momento en el que parecía que todos estaban fascinados pero va cayendo la máscara 🤖☠️
1
21
1,154
No entiendo por qué a la gente le gusta cargar la compra en totebags y bolsas de mercado que nomás te dejan adolorida y son incómodas. A una mochila le cabe todo y es muuucho más cómoda y ergonómica.
1
6
467
El usuario Néstor García compartió ayer, en su cuenta de facebook, este video en el que se muestra una gran cantidad de peces muertos en el Río Santiago, a la altura de San Martín de las Cañas, municipio de Tequila, Jalisco. facebook.com/share/r/1K9jcMK…
20
246
17
339
7,728
Ya me cansé de esa literatura que narra en primera persona lo que sucede afuera como si se tratara de un documental de animales. No hay profundidad, solo descripciones anodinas de problemas actuales. Panfletaria y superficial pero vende y eso es lo que cuenta. Zzz.
17
568
Replying to @gatosyalegatos
Es que le veo mil complicaciones a la hora de hacerlo política pública y me da terror. Veo lo que pasa en Canadá y ufff.
1
61
Replying to @gatosyalegatos
Pero el acompañamiento médico no asegura que sea sin complicaciones, como todo procedimiento médico. No ayudaría igual una política de reducción del daño para quienes igual ya quieren morir?
1
48
Pero eso me dice que entonces se requiere desestigmatizar un tabú. Pero como no podemos o queremos, entonces que entre el Estado?
1
102
Replying to @gatosyalegatos
Lo que no entiendo, salvo en casos en los que la persona no se puede mover, es por qué el suicidio no es una opción digna. Tan indigna es que necesita de una institución pública lo sustituya? Es porque puede fallar? También las inyecciones.
1
1
728
Replying to @ren_villalobos
Pero entonces es un tema de no querer ir en contra de un tabú? Eso no es buen argumento.
1
919
Replying to @gatosyalegatos
No entendí. El suicidio es necesariamente doloroso y poco efectivo?
3
5
1,770