Characterizing When Language Models Take Harmful Actions to Protect Their Assigned Goals
cs.LG, cs.AI, cs.CL
Submitted: 2026-11-02
Updated: 2026-11-02
Terminology
Sources
- Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback
- Taken out of context: On measuring situational awareness in LLMs
- Will AI Tell Lies to Save Sick Children? Litmus-Testing AI Values Prioritization with AIRiskDilemmas
- Alignment faking in large language models
- Empirical Evidence for Alignment Faking in a Small LLM and Prompt-Based Mitigation Techniques
- Frontier Models are Capable of In-context Scheming
- Cognitive models can reveal interpretable value trade-offs in language models
- AgentMisalignment: Measuring the Propensity for Misaligned Behaviour in LLM-Based Agents
- Training language models to follow instructions with human feedback
- Qwen2.5 Technical Report
- $\tau$-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains
- When Preferences Fail to Become Incentives: A Utility-Behavior Gap in Large Language Models
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