Diffuse AI Control on Fuzzy Tasks
cs.LG
Submitted: 2026-06-08
Updated: 2026-11-04
Code: https://github.com/algorithmicsuperintelligence/openevolve
Terminology
Sources
- GEPA: Reflective Prompt Evolution Can Outperform Reinforcement Learning
- Sabotage Evaluations for Frontier Models
- Ctrl-Z: Controlling AI Agents via Resampling
- Measuring Progress on Scalable Oversight for Large Language Models
- An alignment safety case sketch based on debate
- Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision
- Scheming AIs: Will AIs fake alignment during training in order to get power?
- Auditing Sabotage Bench: A Benchmark for Detecting and Fixing Research Sabotage in ML Codebases
- AI Control: Improving Safety Despite Intentional Subversion
- Alignment faking in large language models
- Games for AI Control: Models of Safety Evaluations of AI Deployment Protocols
- ImageDoctor: Diagnosing Text-to-Image Generation via Grounded Image Reasoning
- Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
- AI safety via debate
- Measuring AI Ability to Complete Long Software Tasks
- Combee: Scaling Prompt Learning for Self-Improving Language Model Agents
- Agentic Misalignment: How LLMs Could Be Insider Threats
- True Self-Supervised Novel View Synthesis is Transferable
- Illuminating search spaces by mapping elites
- The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
Related papers
- 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
- (Mis)generalization of Helpful-Only Fine-tuning