CDCK/MoM cita posts como User1, User2 etc. -- Problema de qualidade do LLM?

Falco · Agosto 10, 2026, 4:49pm

Obrigado pelo relatório, este PR trata deste problema:

github.com/discourse/discourse

DEV: Use a realistic few-shot example in the summarizer agent (#42445)

main ← dev-summarizer-realistic-example

merged 06:41PM - 07 Aug 26 UTC

xfalcox

+14 -3

Previously, the summarizer agent's few-shot example used a two-post toy thread w…hose format didn't match the prompts the summarization strategies actually build, and the system prompt described a post format (`N) username message`) that no strategy emits. This change replaces the example with one mirroring a real captured request (discussion title line, posts inside `<input></input>` tags in the real `(N username said: ...` format) paired with a multi-paragraph cited summary demonstrating `[username]({resource_url}/N)` citations, and fixes the post format description to match what the topic, chat, and gist strategies actually send.

O resumo desse tópico agora parece melhor.

Tópico		Respostas	Visualizações
Moderation tool for formatting code with AI Feature ai	10	411	3 de Abril de 2025
Code fences don't seem to deal with C comments Support	26	4796	6 de Outubro de 2015
HTML comments are also summarized by AI Feature ai , ai-summarize	5	299	1 de Julho de 2026
Posts being made as HTML comments by default? Support	3	667	17 de Dezembro de 2020
Pasted content gains erroneous line breaks Bug needs-repro	5	255	19 de Agosto de 2024

CDCK/MoM cita posts como User1, User2 etc. -- Problema de qualidade do LLM?

Tópicos relacionados