False Frontiers: Diagnosing and Mitigating Co-Cheating in Self-Evolving Search Agents

arXiv:2609.39102 · cs.CL, cs.AI, cs.LG · Submitted 2026-09-30 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-09-30

Updated: 2026-10-03

Terminology

• Verification — 11× in this paper · explained in 2 episodes
• Benchmark — 8× in this paper · explained in 2 episodes
• Accuracy — 7× in this paper · explained in 1 episode
• Reinforcement Learning — 7× in this paper · explained in 4 episodes
• Empirical Methods in Natural Language Processing — 6× in this paper
• Pseudo-labels — 6× in this paper · explained in 2 episodes
• Provenance — 5× in this paper · explained in 3 episodes
• Alignment — 3× in this paper · explained in 1 episode
• Natural Questions — 3× in this paper
• Robustness — 3× in this paper · explained in 5 episodes
• Ablation — 2× in this paper · explained in 1 episode
• Trace — 2× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Agentic Search — 1× in this paper · explained in 1 episode
• Consistency — 1× in this paper · explained in 3 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• Few-Shot Learning — 1× in this paper · explained in 4 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• In Second Conference on Language Modeling — 1× in this paper
• Instruction Tuning — 1× in this paper · explained in 2 episodes
• LLM-as-a-Judge — 1× in this paper · explained in 3 episodes
• Pseudo-labeling — 1× in this paper · explained in 3 episodes
• Reward Hacking — 1× in this paper · explained in 8 episodes
• Scaling Laws — 1× in this paper · explained in 4 episodes
• Self-Evolving Agents — 1× in this paper · explained in 2 episodes
• Self-distillation — 1× in this paper · explained in 6 episodes
• Semi-supervised Learning — 1× in this paper · explained in 4 episodes
• Sycophancy — 1× in this paper · explained in 4 episodes

Sources

• Concrete Problems in AI Safety
• Self-Questioning Language Models
• REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
• R-Zero: Self-Evolving Reasoning LLM from Zero Data
• WebSailor: Navigating Super-human Reasoning for Web Agent
• OpenResearcher: A Fully Open Pipeline for Long-Horizon Deep Research Trajectory Synthesis
• Search-E1: Self-Distillation Drives Self-Evolution in Search-Augmented Reasoning
• SPICE: Self-Play In Corpus Environments Improves Reasoning
• CAFE: Self-Improving Search Agents Need Co-Evolving Feedback
• WebExplorer: Explore and Evolve for Training Long-Horizon Web Agents
• WebGPT: Browser-assisted question-answering with human feedback
• Asymmetric self-play for automatic goal discovery in robotic manipulation
• R1-Searcher: Incentivizing the Search Capability in LLMs via Reinforcement Learning
• ZeroSearch: Incentivize the Search Capability of LLMs without Searching
• SearchMaster: Grounded and Regulated Self-Play for Search Agents
• Socratic-Zero : Bootstrapping Reasoning via Data-Free Agent Co-evolution
• StepSearch: Igniting LLMs Search Ability via Step-Wise Proximal Policy Optimization
• Corrective Retrieval Augmented Generation
• Scalable Reinforcement Post-Training Beyond Static Human Prompts: Evolving Alignment via Asymmetric Self-Play
• Dr. Zero: Self-Evolving Search Agents without Training Data

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