QRAE: Query-Relevant Attention Entropy for Context Compression in DeepSearch Agents
cs.LG
Submitted: 2026-11-05
Updated: 2026-11-05
Terminology
Sources
- xbench: Tracking Agents Productivity Scaling with Profession-Aligned Real-World Evaluations
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- SideQuest: Model-Driven KV Cache Management for Long-Horizon Agentic Reasoning
- Fact, Fetch, and Reason: A Unified Evaluation of Retrieval-Augmented Generation
- IntentKV: Cross-Turn Intent-Aware KV Cache Pruning for Agent Inference
- LiteResearcher: A Scalable Agentic RL Training Framework for Deep Research Agent
- OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
- Towards Efficient Agents: A Co-Design of Inference Architecture and System
- Scaling Long-Horizon LLM Agent via Context-Folding
- Tongyi DeepResearch Technical Report
- DuoAttention: Efficient Long-Context LLM Inference with Retrieval and Streaming Heads
- AgentFold: Long-Horizon Web Agents with Proactive Context Management
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- FlexMoE: Learning One-for-All Multi-Budget Expert Pruning for Flexible MoE Deployment
- OptiMind: Improving Small Language Models for Optimization Modeling through Formulation Guidance and Multi-Turn Learning
- Staying in the Sweet Spot: Boost Reasoning Evolution via Responsive Rollout Difficulty Manipulation
- Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
- Rethinking Black-Box On-Policy Distillation: A Rubric-based Analysis