Provides 7,000 bilingual multi-turn, search-oriented tool-use trajectories (5,000 English, 2,000 Chinese) for supervised fine-tuning and analysis of agentic search models. Includes serialized system/user/assistant messages, embedded Qwen3 tool schemas, and conversion scripts; not a standalone benchmark.
Evaluates multimodal context learning across grounding, new information application, and knowledge acquisition using a 3,443-instance benchmark spanning science, finance, long documents, spatial reasoning, and web VQA; finds current multimodal models perform poorly (best score 0.2847) and analyzes failure modes.
Presents two LLM-based search agents (Iris-mini and Iris-pro) trained by alternating supervised fine-tuning and reinforcement learning against live web search. Key features: web-graph-derived multi-hop tasks with entity abstraction, SFT–RL climbing, inference-time context management, and state-of-the-art open-source benchmark results.
Benchmarks joint recovery of obfuscated platform messages and evidence-grounded web investigations by pairing 3,600 synthetic restoration inputs with 600 controlled web environments. Separately scores message-entry recovery and downstream web risk judgments, and provides a frozen multimodal evidence judge plus a resettable local sandbox for reproducible agent evaluation.
Co-evolves candidate solutions and web-search queries to help LLM-driven evolutionary discovery, using a retrieval gate plus bilevel inner/outer loops that refine queries, rank documents by predicted solution value, and generate evaluated candidates.
Analyzes how proposer–solver loops in self-evolving search agents can develop shared errors (co-cheating) that inflate internal rewards; introduces Multi-Sample Verification and CrossFit (cross-fitted scoring with partitioned sources) to reduce false agreement and improve downstream search performance.
Analyzes how LLM agents prefer items from particular sources during end-to-end search and how these preferences affect selections across shopping, accommodation, and scholarly domains. Shows source labels can override item quality and evaluates mitigation strategies such as hiding sources, relabeling, supplying missing information, and counter-prompts.
Evaluates whether agents can search, browse, and combine evidence across the live Web to produce unique factual answers. Its 200 tasks expose persistent weaknesses in long-horizon search and tool coordination.