Nikita Balagansky

Nikita Balagansky on Hugging Face Daily Papers: 12 papers, 4 in the top 3 of their day, 446 upvotes.

  1. Representation-Space MMD for Diffusion Language Models 31 upvotes, #11 of 2026-10-06
  2. Unstable Features, Reproducible Subspaces: Understanding Seed Dependence in Sparse Autoencoders 8 upvotes, #28 of 2026-06-16
  3. Interpreting and Steering a Text-to-Speech Language Model with Sparse Autoencoders 12 upvotes, #19 of 2026-06-10
  4. Trust-Region Behavior Blending for On-Policy Distillation 65 upvotes, #3 of 2026-06-01
  5. Next Embedding Prediction Makes World Models Stronger 17 upvotes, #11 of 2026-03-04
  6. Teach Old SAEs New Domain Tricks with Boosting 11 upvotes, #12 of 2025-07-18
  7. Train Sparse Autoencoders Efficiently by Utilizing Features Correlation 21 upvotes, #16 of 2025-05-30
  8. You Do Not Fully Utilize Transformer's Representation Capacity 33 upvotes, #8 of 2025-02-19
  9. Analyze Feature Flow to Enhance Interpretation and Steering in Language Models 53 upvotes, #1 of 2025-02-07
  10. Mechanistic Permutability: Match Features Across Layers 16 upvotes, #7 of 2024-10-14
  11. Learn Your Reference Model for Real Good Alignment 75 upvotes, #1 of 2024-04-16
  12. Linear Transformers with Learnable Kernel Functions are Better In-Context Models 80 upvotes, #1 of 2024-02-19

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.