IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis

arXiv:2609.29444 · cs.CL, cs.AI · Submitted 2026-09-24 · Read on arXiv

cs.CL, cs.AI

Submitted: 2026-09-24

Updated: 2026-09-24

Code: https://github.com/Tencent/IterSynth

Terminology

• Benchmark — 42× in this paper · explained in 2 episodes
• Accuracy — 25× in this paper · explained in 1 episode
• Credit assignment — 12× in this paper · explained in 3 episodes
• Reinforcement Learning — 9× in this paper · explained in 4 episodes
• Role-Decoupled Policy Optimization — 9× in this paper
• Fine-Tuning — 7× in this paper · explained in 5 episodes
• Perplexity — 7× in this paper · explained in 3 episodes
• Foundation Model — 6× in this paper · explained in 3 episodes
• GRPO — 6× in this paper · explained in 6 episodes
• Supervised Fine-Tuning — 6× in this paper
• Decoupling — 5× in this paper · explained in 2 episodes
• Verification — 5× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Foundation Models — 4× in this paper · explained in 7 episodes
• LLM judge — 4× in this paper · explained in 2 episodes
• Action Space — 3× in this paper · explained in 1 episode
• Supervised Fine-Tuning (SFT) — 3× in this paper · explained in 8 episodes
• Trace — 3× in this paper · explained in 2 episodes
• Consistency — 2× in this paper · explained in 3 episodes
• Multi-Agent System — 2× in this paper · explained in 6 episodes
• Test-time Scaling — 2× in this paper · explained in 2 episodes
• Aggregation — 1× in this paper · explained in 1 episode
• Alternation — 1× in this paper · explained in 1 episode
• Benchmarking — 1× in this paper · explained in 3 episodes
• Entanglement — 1× in this paper · explained in 4 episodes
• Failure Modes — 1× in this paper · explained in 3 episodes
• Knowledge Graph — 1× in this paper · explained in 7 episodes
• Knowledge graphs — 1× in this paper · explained in 2 episodes
• Model-Agnostic — 1× in this paper · explained in 2 episodes
• Reasoning Traces — 1× in this paper · explained in 3 episodes
• Reward Hacking — 1× in this paper · explained in 8 episodes
• Robustness — 1× in this paper · explained in 5 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Two-stage training — 1× in this paper · explained in 2 episodes

Sources

• AgentCPM-Explore: Realizing Long-Horizon Deep Exploration for Edge-Scale Agents
• REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
• OpenSeeker: Democratizing Frontier Search Agents by Fully Open-Sourcing Training Data
• Kimi K2.5: Visual Agentic Intelligence
• WebSailor-V2: Bridging the Chasm to Proprietary Agents via Synthetic Data and Scalable Reinforcement Learning
• WebWeaver: Structuring Web-Scale Evidence with Dynamic Outlines for Open-Ended Deep Research
• WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
• InfoFlow: Reinforcing Search Agent Via Reward Density Optimization
• GAIA: a benchmark for General AI Assistants
• MiroThinker: Pushing the Performance Boundaries of Open-Source Research Agents via Model, Context, and Interactive Scaling
• Toolformer: Language Models Can Teach Themselves to Use Tools
• HybridFlow: A Flexible and Efficient RLHF Framework
• WebShaper: Agentically Data Synthesizing via Information-Seeking Formalization
• BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
• ReSum: Unlocking Long-Horizon Search Intelligence via Context Summarization
• Qwen3 Technical Report
• ReAct: Synergizing Reasoning and Acting in Language Models
• AgentFold: Long-Horizon Web Agents with Proactive Context Management
• MemAgent: Reshaping Long-Context LLM with Multi-Conv RL-based Memory Agent
• InfoAgent: Advancing Autonomous Information-Seeking Agents

Related papers

• Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
• Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
• Subliminal Steering: Stronger Encoding of Hidden Signals
• MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
• The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
• Untangling the Mechanisms of Misleading Context in Medical Question Answering