Instrumental Monitor Evasion Emerges Under Ordinary Task Pressure
ExplorationBench: Measuring AI Systems' Exploration in Verifiable Alien Worlds
VSCode's SSH Agent Is Bananas
2 min1
IterSynth: Rethinking Deep Search Agents via Role-Decoupled Iterative Synthesis
Agent-Editing World Model: Rethinking World Modeling for LLM Agents
Just-in-Time Memory: Learning to Curate Task-Adaptive Memory for LLM Agents
Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents
Rouxii: Exploiting Honeypots with Deception-Aware AI Pentesters
Making Agents More Consistent: Skills Should Form Habits for Repeat Tasks
RRSI: Regularized Recursive Self-Improvement of Agent Harnesses
DolphinBench: Mapping the Pareto Frontier of Agent Memory
Emergent Collusion in Long-Horizon LLM Agent Interaction
Jev for Scientific Decisions: Evaluating Semantic Choices and Their Consequences
ActGov: Governing LLM Agent Actions via Policy-Constrained Validation
1
LeaseGuard: Incumbent-Preserving Admission Control for Privileged LLM Agents
Jev-Mem: System-One-Controlled Agentic Memory for Efficient AI Agents
An Interpretable Memory Decision Controller for LLM Agents Based on Three-Signal Complementarity: Decoupling Confidence and Consistency
Bayesian Belief Layer for Controllable Opinion Dynamics in LLM Agents
CIPL: A Channel-Aware Framework for Recoverable Privacy Leakage in LLM Agents
1
Could AI really kill us all? Your questions, answered.
8 min
Best Open-Source Agent Harnesses for Local LLMs in 2026
7 min1
GraphSkillEvo: Evolutionary Optimization of Graph-Structured Agent Skills
