On the Monitorability of Natural Language Autoencoders: Identification vs Discrimination
cs.LG, cs.CL
Submitted: 2026-10-13
Updated: 2026-10-21
Comments: work in progress, open draft
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Analyzing and Improving Chain-of-Thought Monitorability Through Information Theory
- A Decision-Theoretic Formalisation of Steganography With Applications to LLM Monitoring
- Decoding Hidden Deception in Reasoning LLMs: Activation Explainers for Deception Auditing
- "Did you lie?" Evaluating Lie Detectors across Model Scale and Belief-Verified Model Organisms
- Eliciting Secret Knowledge from Language Models
- ReferentialGym: A Nomenclature and Framework for Language Emergence & Grounding in (Visual) Referential Games
- Off-Belief Learning
- Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
- Liars' Bench: Evaluating Lie Detectors for Language Models
- Emergent Multi-Agent Communication in the Deep Learning Era
- LLMs Can Covertly Sandbag on Capability Evaluations Against Chain-of-Thought Monitoring
- Do Activation Verbalization Methods Convey Privileged Information?
- ResearchArena: Evaluating Sabotage and Monitoring in Automated AI R&D
- On the Pitfalls of Measuring Emergent Communication
- Eliciting Latent Knowledge from Quirky Language Models
- Auditing language models for hidden objectives
- Emergent Communication of Generalizations
- One Probe Won't Catch Them All: Towards Targeted Deception Detection
- Benchmarking Deception Probes via Black-to-White Performance Boosts
- Qwen2.5 Technical Report
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks