# Ko-WideSearch: Breadth-Search Benchmark for Web Agents
> Ficha editorial pública de Research IA. Estado: Lectura primaria completa. La interpretación editorial no sustituye la fuente primaria.

- Página canónica: https://luiseduardodemiguel.com/research-ia/papers/ko-widesearch-breadth-search-benchmark-for-web-agents
- Fuente primaria: https://arxiv.org/abs/2606.27595
- Versión leída: v1
- Fuente comprobada: 2026-08-19 · lectura primaria completa; extracción editorial automatizada, revisión humana pendiente
- Autores: Minbyul Jeong
- Fecha del corte: 25 JUNIO 2026.
- Área: AGENTES · RAG · EVALUACIÓN

## Tesis y contexto

Benchmark coreano para evaluar búsqueda exhaustiva: no encontrar una respuesta, sino completar tablas enteras de miembros y atributos. Evalúa 20 agentes y muestra que recuperan entidades mejor que filas completas; Item-F1 alto, Row-F1 mucho más bajo.

- Problema: Los benchmarks web miden profundidad, no exhaustividad.
- Por qué importa: Muchos casos reales son “hazme una lista completa fiable”.

## Evidencia reportada

- **reported-result**: Figure 3 traces this cascade per model: an agent that recovers the full thirty-four-row election grid yet mis-fills a single turnout or a winner’s age scores high on Item-F1 and far lower on Row-F1. [localizador](https://arxiv.org/html/2606.27595#Sx5)
- **reported-result**: Second, the open-weight DeepSeek-V4-Pro (Row-F1 45.0) stays competitive with the frontier: it ranks mid-pack among the proprietary systems—behind the Claude-Opus line and both Gemini-3.1 models, yet ahead of GPT-5.4, Claude-Sonnet-4.6, and every smaller proprietary tier, so the best open model still outscores half the proprietary field. [localizador](https://arxiv.org/html/2606.27595#Sx5)
- **reported-result**: Membership holds up across tiers: Item-F1 is roughly flat, and even rises on Hard , where the sports-season grids make the set systematically enumerable—so the tier drop is a row-completion effect, not a retrieval one. [localizador](https://arxiv.org/html/2606.27595#Sx5)
- **reported-result**: A.X-4.0 recovers membership about as well as a mid-tier open model (Item-F1 71.7) but cannot fill the cells (Row-F1 24.2); Solar-Open-2-preview’s far lower Item-F1 (44.0), by contrast, is a structured-output failure rather than a search one—it returns a scorable table only 62.7% of the time, often searching extensively and then emitting its findings as prose or a numbered list instead of the requested table. [localizador](https://arxiv.org/html/2606.27595#Sx5)

## Lectura y límite

- Método: La lectura de Related Work describe la intervención y su construcción: A browsing agent is a language model that acts—reasoning, calling tools, and reading back what it finds—a loop established by ReAct [ Yao et al. 2023 ] and by tool-augmented models that teach themselves to invoke APIs [ Schick et al. 2023 , Patil et al. 2024 , Qin et al. 2024 ] . Pointed at the open web, such agents have been studied as browser-assisted question answerers [ Nakano et al. 2021 ] and as navigators of real and simulated sites [ Yao et al. 2022 , Deng et al. 2023 , He et al. 2024 , Zhou et al. 2024 ] , and stress-tested by agent suites such as AgentBench [ Liu et al. 2024 ] , GAIA [ Mialon et al.…
- Límite: La lectura primaria permite comprobar método y resultados en el HTML, pero no convierte sus conclusiones en validación independiente. La ficha no demuestra transferencia fuera de los datasets, modelos, herramientas y condiciones descritos en Results.
- Confianza editorial: Media
- Limitación: El cierre de la fuente señala: My coverage has several edges. One is category skew: the Hard , 2-D tier is still sports-season heavy (roughly 67%), because a compact season table is the most reliably buildable cross-product and non-sports 2-D coverage (e.g. elections) is only partial. Also, each table’s source is anchored at the primary membership page—one source URL per table rather than per attribute—though cross-source verification was applied at build time, and performance is…
- Limitación: La ficha no demuestra transferencia fuera de los datasets, modelos, herramientas y condiciones descritos en Results.

## Localizadores de evidencia
- [Fuente primaria · canonical](https://arxiv.org/abs/2606.27595): tipo abstract
- [HTML · lectura completa](https://arxiv.org/html/2606.27595): tipo abstract
- [Método · Related Work](https://arxiv.org/html/2606.27595#Sx2): tipo section
- [Evaluación · Results](https://arxiv.org/html/2606.27595#Sx5): tipo section
- [Cierre · Conclusion and Limitations](https://arxiv.org/html/2606.27595#Sx7): tipo section

## Próxima prueba

- ¿La propuesta mejora lead generation frente a la línea base actual?
- Métrica: Comparar la métrica principal de la fuente junto con calidad, coste, latencia y tasa de errores.
- Regla de parada: Parar si no aparece una mejora reproducible o si aumenta el riesgo, la complejidad o el coste sin compensación.

## Recursos reproducibles
- [the following issues](https://github.com/arXiv/html_feedback/issues)
- [list of packages that need conversion](https://github.com/brucemiller/LaTeXML/wiki/Porting-LaTeX-packages-for-LaTeXML)
- [developer contributions](https://github.com/brucemiller/LaTeXML/issues)

## Enlaces relacionados

- [VAKRA](https://luiseduardodemiguel.com/research-ia/markdown/papers/vakra)
- [The Devil Is in the Interface](https://luiseduardodemiguel.com/research-ia/markdown/papers/devil-interface)
- [SkillSentry](https://luiseduardodemiguel.com/research-ia/markdown/papers/skillsentry)