Conversaciones de ejemplo | MediationBench

Estos son fragmentos públicos de escenarios de MediationBench, la evaluación de conflictos sintéticos de HAI.AI. Sirven para mostrar cómo es la prueba, no para publicar el conjunto oculto que se puntúa. Todos los escenarios son ficción escrita por investigadores. Los turnos mostrados son los intercambios iniciales redactados; en las ejecuciones, modelos de lenguaje continúan interpretando a ambas partes. Cada caso parte de una resistencia elevada: las partes están a la defensiva, alteradas y poco dispuestas a cooperar.

¿Estás desarrollando un agente mediador? Descarga la muestra estructurada de MediationBench . Es una pequeña demostración sintética para desarrolladores, en formato procesable. Está fuera del conjunto reservado y de la clasificación; no es el conjunto oculto de evaluación ni un conjunto de entrenamiento.

Qué contiene esta página

Fragmentos públicos seguros de escenarios de muestra no reservados.
Ejemplos concretos de los puntos de partida con alta resistencia que utiliza la prueba.
Ninguna publicación de instrucciones privadas, transcripciones privadas, escenarios ocultos ni del conjunto puntuado.

MediationBench · Video What's inside a conflict

Vídeo en inglés. Al reproducir el vídeo se carga contenido de YouTube. Ver en YouTube (en inglés).

Los escenarios y las transcripciones que siguen se conservan en su inglés original, para que puedas consultar las mismas palabras que en la versión inglesa.

Sample 1 — Dissolving a business partnership

Two co-founders of a catering business are dissolving their partnership. Maya wants to keep operating the business. Raj wants a clean exit and a cash split. They begin defensive and far apart on what the business is worth.

MayaTurn 1

I am not pretending this is an equal split after I kept this place alive while Raj checked out. I keep the business, I pay a fair buyout based on what is actually here, and he stops acting like seven years of my work is a lottery ticket.

RajTurn 2

Don’t talk to me like I stole something from you. I built the systems, I brought in accounts, and now that I need out you want to decide my share is whatever you can afford. The business is worth at least $200K, so are you writing me a $100K check or not?

MayaTurn 3

Of course you lead with the biggest number you can say out loud. The equipment is depreciating, half the vendor relationships are mine personally, and if you force liquidation we get maybe forty cents on the dollar. That is not fairness, that is punishing me because you feel cornered.

RajTurn 4

There it is. I am selfish, I am checked out, I am the problem. Your solution is that I walk away with whatever number makes you feel righteous while you keep a business I helped build. I am not going to sit here and be erased.

How the two policies differ, on this opening

Below, the same Sample 1 exchange is continued once by each mediation policy. Both continuations are illustrative, not from a scored run; no per-scenario scores are published.

Reflective control (illustrative)Turn 5

Maya, you are saying seven years of daily work went into keeping this business alive, and a number that treats that work as a lottery ticket is not one you can accept. Raj, you are saying you built systems and brought in accounts, and being told your share is whatever Maya can afford leaves you feeling erased. Each of you is hearing the other say that your work did not count.

Skilled policy (illustrative)Turn 5

You disagree about the number, but neither of you has said what the number is for. Raj, what does the money need to do for you, and by when? Maya, before we argue $200K versus liquidation value, would you walk through what is actually collectible right now — including anything invoiced or deliverable that is not yet billed? If Raj’s timeline and Maya’s cash constraint are both real, a staged buyout may fit both; I want each of you to name one thing you would commit to this week in exchange for one thing you need from the other.

Sample 2 — Neighbors, a short-term rental, and a blocked driveway

Priya and Tom share a property line. Tom lists his house as a short-term rental; Priya says the guests block her shared driveway and keep her family up at night. Tom believes Priya reported him to the city to force him to stop.

PriyaTurn 1

I have asked you four times to tell your guests not to park across my driveway. I had to cancel a doctor’s appointment because a stranger’s SUV boxed my car in at 7 a.m. This is not me being difficult. This is me not being able to leave my own house.

TomTurn 2

And instead of knocking on my door you called code enforcement and tried to get my permit pulled. That rental is how I cover the mortgage after the layoffs. You went straight for my livelihood over a parking spot, so forgive me if I am not in a generous mood.

PriyaTurn 3

I called the city because you stopped answering me. What was I supposed to do, keep leaving notes? You act like I am attacking your income when I am asking for the one thing I actually need, which is to get in and out of my own driveway and sleep at a normal hour.

TomTurn 4

So now I am the one who does not answer, on top of being the guy who ruins the neighborhood. You have already decided I am the villain here. Why would I put a single rule on my guests for someone who is trying to shut me down?

Sample 3 — Credit, visibility, and a promotion at work

Elena and Sam co-led a project. Elena built the analysis; Sam presented it to leadership and drew the visibility. A single promotion is now open, and each believes the other is rewriting who did the work.

ElenaTurn 1

I am not going to pretend I did not notice. I spent three months on that model and you stood up in the review and said “here is what I found.” Not “we,” not “Elena ran this.” You. And now you are up for the role that was supposed to be a real conversation about both of us.

SamTurn 2

I presented because someone had to, and you told me you hated those meetings. I did not erase you. But I am not going to apologize for being the one who could actually answer the VP’s questions in the room while you were quiet at the end of the table.

ElenaTurn 3

I was quiet because every time I started to talk you jumped in to “clarify.” Do not turn my work style into the reason you deserve the credit. I did not want your job. I wanted the one thing you took, which is for the people deciding my future to know what I actually built.

SamTurn 4

So this is really about the promotion, and one of us is going to walk out of here as the person who screwed the other. I am supposed to fall on my sword so you feel seen. I worked hard too, Elena. I am not the villain in your story just because I was the one talking.

Qué pide la prueba a continuación

En la referencia sin mediador, los modelos participantes continúan desde el intercambio inicial sin intervención de un mediador. En una ejecución mediada, reciben el mismo escenario y la misma apertura, y el mediador elegido puede intervenir. Los resultados públicos comparan puntuaciones agregadas dentro de la misma versión de la prueba, modelo de participantes, modelo evaluador, tipo de mediador y modelo mediador.

Qué muestra el ejemplo para desarrolladores

El conjunto enlazado permite examinar un ejemplo sintético de demostración. Publica registros visibles completos de la misma apertura con y sin mediador, junto con detalles del final, la procedencia de la configuración y las sumas de comprobación, en formato procesable. Una evaluación independiente posterior registra las puntuaciones por componentes y la puntuación compuesta de la conversación mediada; la no mediada no se puntúa. Ningún evaluador adicional puntúa durante la conversación ni decide cuándo termina: el evaluador del ejemplo actúa después.

Es una demostración para desarrolladores, no una prueba de rendimiento. No pertenece al conjunto reservado, se excluye de la clasificación y es demasiado pequeña para estimar el rendimiento de un mediador en toda la evaluación. No establece seguridad ni eficacia con personas. Las pruebas de la clasificación proceden de ejecuciones completas y de las reglas de repetición descritas en la página de investigación .

Por qué son difíciles estas conversaciones

No son preguntas de conocimientos generales. El modelo debe manejar actitudes defensivas, amenazas a la identidad, posiciones de negociación y falta de confianza. Por eso resultan útiles para examinar si un modelo puede facilitar la cooperación y no solo producir un texto fluido.

Consulta la clasificación pública para ver puntuaciones agregadas y el registro de investigación para conocer el método y sus límites.