Can an AI Agent Be Epistemically Akratic? Higher-Order Evidence & the Limits of Self-Correction
cs.AI
Submitted: 2026-11-11
Updated: 2026-11-11
Terminology
Sources
- Does Self-Evaluation Enable Wireheading in Language Models?
- Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
- Looking Inward: Language Models Can Learn About Themselves by Introspection
- Do LLM Evaluators Prefer Themselves for a Reason?
- Challenges with unsupervised LLM knowledge discovery
- Language Models (Mostly) Know What They Know
- How Overconfidence in Initial Choices and Underconfidence Under Criticism Modulate Change of Mind in Large Language Models
- Causal Evidence that Language Models use Confidence to Drive Behavior
- Are You Sure? Challenging LLMs Leads to Performance Drops in The FlipFlop Experiment
- LLMs Know More Than They Show: On the Intrinsic Representation of LLM Hallucinations
- Can Large Reasoning Models Self-Train?
- Are LLM Decisions Faithful to Verbal Confidence?
Related papers
- Multi-Objective Structured Pruning of LLMs for Latency and Model Size Optimization
- 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
- Embeddings for Preferences, Not Semantics
- Can We Trust Item Response Theory for AI Evaluation?
- Diagnosing Task Insensitivity in Language Agents