User Embeddings are Superpositions of Interpretable Behavioral Modes
cs.LG
Submitted: 2026-11-05
Updated: 2026-11-07
Terminology
Sources
- An X-Ray Is Worth 15 Features: Sparse Autoencoders for Interpretable Radiology Report Generation
- Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
- A is for Absorption: Studying Feature Splitting and Absorption in Sparse Autoencoders
- Modeling User Behaviour in Research Paper Recommendation System
- Persona Vectors: Monitoring and Controlling Character Traits in Language Models
- CytoSAE: Interpretable Cell Embeddings for Hematology
- Toy Models of Superposition
- HiD-VAE: Interpretable Generative Recommendation via Hierarchical and Disentangled Semantic IDs
- Scaling and evaluating sparse autoencoders
- Sparse Autoencoders for Sequential Recommendation Models: Interpretation and Flexible Control
- Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?
- Learning biologically relevant features in a pathology foundation model using sparse autoencoders
- The Assistant Axis: Situating and Stabilizing the Default Persona of Language Models
- Item Recommendation with Evolving User Preferences and Experience
- User-LLM: Efficient LLM Contextualization with User Embeddings
- Decoding Dense Embeddings: Sparse Autoencoders for Interpreting and Discretizing Dense Retrieval
- Finding Lookalike Customers for E-Commerce Marketing
- Sparse Autoencoder Insights on Voice Embeddings
- Improving Dictionary Learning with Gated Sparse Autoencoders
- A Comprehensive Review of Recommender Systems: Transitioning from Theory to Practice
Related papers
- A Physics-Regulated Neural Framework for Learning 3D Grain Growth Dynamics
- 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