Verbalizing Subliminal Learning Effects Using Text Optimization
cs.LG, cs.AI, cs.CL
Submitted: 2026-09-15
Updated: 2026-09-15
Code: https://github.com/nathanhu0/verbalizing-subliminal-learning
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Subliminal Effects in Your Data: A General Mechanism via Log-Linearity
- A General Language Assistant as a Laboratory for Alignment
- Dataset distillation for memorized data: Soft labels can leak held-out teacher knowledge
- Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
- Subliminal Learning Is Steering Vector Distillation
- Sycophantic Agreement Transfers with Neutral Data via Contrastive Preference Optimization
- SelfIE: Self-Interpretation of Large Language Model Embeddings
- Subliminal Learning: Language models transmit behavioral traits via hidden signals in data
- Phantom Transfer: Data Poisoning can Survive Data-Level Defences
- Detecting Adversarial Fine-tuning with Auditing Agents
- Geometric-Averaged Preference Optimization for Soft Preference Labels
- Learning to Interpret Weight Differences in Language Models
- ContextBench: Modifying Contexts for Targeted Latent Activation
- Gradient-based Adversarial Attacks against Text Transformers
- Covert Malicious Finetuning: Challenges in Safeguarding LLM Adaptation
- LoRA: Low-Rank Adaptation of Large Language Models
- Categorical Reparameterization with Gumbel-Softmax
- Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
- The Power of Scale for Parameter-Efficient Prompt Tuning
- LARGO: Latent Adversarial Reflection through Gradient Optimization for Jailbreaking LLMs
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: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks