# VibeSearchBench: Benchmarking Long-horizon Proactive Search in the Wild

arXiv: https://arxiv.org/abs/2605.27882 · HTML: https://mlsanity.com/paper/2605.27882/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Web, search, and research agents, one of 16 directions in LLM agents and tool use.
- Builds on 8 of its 21 references that are in the map.
- First posted on arXiv 27 May 2026.

## Placed in

- LLM agents and tool use / [Web, search, and research agents](https://mlsanity.com/agents/web-search-and-research-agents/)

## Builds on (most cited first, 8 of 8)

- [DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence](https://mlsanity.com/paper/2606.19348/) (2026)
- [BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents](https://mlsanity.com/paper/2504.12516/) (2025)
- [DeepResearch Bench: A Comprehensive Benchmark for Deep Research Agents](https://mlsanity.com/paper/2506.11763/) (2025)
- [ResearchRubrics: A Benchmark of Prompts and Rubrics For Evaluating Deep Research Agents](https://mlsanity.com/paper/2511.07685/) (2025)
- [DeepSearchQA: Bridging the Comprehensiveness Gap for Deep Research Agents](https://mlsanity.com/paper/2601.20975/) (2026)
- [ClawMark: A Living-World Benchmark for Multi-Turn, Multi-Day, Multimodal Coworker Agents](https://mlsanity.com/paper/2604.23781/) (2026)
- [OneMillion-Bench: How Far are Language Agents from Human Experts?](https://mlsanity.com/paper/2603.07980/) (2026)
- [LLM-WikiRace Benchmark: How Far Can LLMs Plan over Real-World Knowledge Graphs?](https://mlsanity.com/paper/2602.16902/) (2026)

## Built on by (most cited first, 1 of 1)

- [Asking for What Was Never Requested: Horizontal and Vertical Proactivity in Agents](https://mlsanity.com/paper/2609.37236/) (2026)

## Abstract (arXiv)

LLM-based agents score well on search benchmarks, yet real users consistently find results unsatisfying, revealing a persistent evaluation-experience gap. We attribute this gap to existing benchmarks' reliance on over-specified queries, single-turn interactions, and fixed-schema evaluation, none of which reflect real search behavior where users and agents collaboratively refine vague intent through multi-turn dialogue. We term this paradigm VibeSearch and introduce VibeSearchBench, a benchmark comprising 200 manually curated bilingual (Chinese and English) tasks across 20 domains, split into VibeSearch-Pro (professional) and VibeSearch-Daily (daily-life) subsets. Each task pairs a user persona with a schema-free ground-truth knowledge graph, and is evaluated through a progressive-disclosure user simulator and a graph-matching evaluation framework. We benchmark seven frontier models under both the ReAct framework and the OpenClaw agent harness. Results show that all models remain substantially inadequate for VibeSearch (best F1: 30.30), highlighting the need for fundamental advances in long-context reasoning, proactive intent elicitation, and structured knowledge construction.