Represented but Not Used: Value-Policy Dissociation in LLM Agents
cs.AI
Submitted: 2026-11-07
Updated: 2026-11-07
Terminology
Sources
- gpt-oss-120b & gpt-oss-20b Model Card
- Understanding intermediate layers using linear classifier probes
- Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States
- Evaluating Long-Context Reasoning in LLM-Based WebAgents
- Is Your LLM Secretly a World Model of the Internet? Model-Based Planning for Web Agents
- World Models
- Causal Mediation Analysis for Interpreting Neural NLP: The Case of Gender Bias
- Interpreting and Steering for Safe and Correct Code Generation
- Qwen2.5 Technical Report
- Representation Engineering: A Top-Down Approach to AI Transparency
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