hermes-ai.net

Leer docs →
HermesHermes Agent Docs
← Back to case library
Case / 2100012553065644103Código
Case media / 1MP4 ↗

Yuntian Deng

@yuntiandeng
Verified

Localized reading

Cada vez creo más que el software del futuro combinará código para el flujo de control con pequeños programas neuronales para emitir juicios «difusos». Eso es lo que exploro con ProgramAsWeights, que responde de dónde salen esos programas neuronales: se «compilan» a partir de descripciones en inglés. Por ejemplo, combiné 30 programas neuronales con un árbol de decisión para crear un asistente para una web de cursos que parece un chatbot, pero funciona localmente. Cada pequeño programa resuelve preguntas como «¿a qué especialista debe derivarse esto?», mientras el código controla todo el flujo…

Original post / EN

I increasingly think future software will combine code for control flow with small neural programs for "fuzzy" judgments. That's what I've been exploring with ProgramAsWeights, which answers the question of where those neural programs come from: they are "compiled" from English descriptions. For example, I combined 30 neural programs with a decision tree to build a course website helper that looks like a chatbot but runs locally. Each small program handles a question like "Which specialist answerer should this be routed to", and code controls the overall flow. You can try building with it here: https://t.co/Bwgcn362bc
views
13,5 mil
likes
171
saves
140
reposts
21
Ver original ↗

Links and projects

Quoted post

Diogo Almeida @CompleteSkeptic

After co-inventing ChatGPT, I kept asking myself: why have superhuman chat models not led to AGI? I’ve spent the last 2 years in stealth building a new way to train models (RLCD), and a new type of frontier AI model that we are releasing today: Jev • 20-200x faster • 40-400x https://t.co/JSybNG2BKJ

Ver original ↗

Fuente

This independent archive preserves a public post with attribution. Text, media, account details and trademarks belong to their respective owners.