BrowseComp-Plus

BrowseComp-Plus BrowseComp-Plus is a new benchmark for Deep-Research system, isolating the effect of the retriever and the LLM agent to enable fair, transparent comparisons of Deep-Research agents. The benchmark sources challenging, reasoning-intensive queries from OpenAI's BrowseComp. However, instead of searching the

Papers using BrowseComp-Plus (9)

FineVerify: Scaling Test-Time Compute with Fine-Grained Self-Verification for Agentic Search2026LACUNA: Safe Agents as Recursive Program Holes2026WebSwarm: Recursive Multi-Agent Orchestration for Deep-and-Wide Web Search2026ActiveMem: Distributed Active Memory for Long-Horizon LLM Reasoning2026LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception2026ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL2026Dr-DCI: Scaling Direct Corpus Interaction via Dynamic Workspace Expansion2026Natural Language Query to Configuration for Retrieval Agents2026SkyRL-Agent: Efficient RL Training for Multi-turn LLM Agent2025