CAFE: Self-Improving Search Agents Need Co-Evolving Feedback
cs.AI
Submitted: 2026-08-25
Updated: 2026-09-15
Code: https://github.com/perplexityai/wandr
Terminology
Sources
- Off-Policy Corrected Reward Modeling for Reinforcement Learning from Human Feedback
- BrowseComp-Plus: A More Fair and Transparent Evaluation Benchmark of Deep-Research Agent
- Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
- Group-in-Group Policy Optimization for LLM Agent Training
- Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learning
- ReSeek: A Self-Correcting Framework for Search Agents with Instructive Rewards
- EvoLM: Self-Evolving Language Models through Co-Evolved Discriminative Rubrics
- InfoFlow: Reinforcing Search Agent Via Reward Density Optimization
- TRAJDEBUG: Tracing Error Lifecycle to Identify Critical Failures in Long-Horizon Agent Trajectories
- Qwen2.5 Technical Report
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- OpenAI GPT-5 System Card
- R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning
- Kimi K2: Open Agentic Intelligence
- Kimi K2.5: Visual Agentic Intelligence
- Solving math word problems with process- and outcome-based feedback
- Information Gain-based Policy Optimization: A Simple and Effective Approach for Multi-Turn Search Agents
- The Long-Horizon Task Mirage? Diagnosing Where and Why Agentic Systems Break
- Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
- BrowseComp: A Simple Yet Challenging Benchmark for Browsing Agents
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection