PPaperPicks

Simone Scardapane

8 papers at tracked venues · 6 at CORE A* · active 20242026

Venues

Frequent coauthors

Maximo Eduardo Rulli
×1
Antonio Andrea Gargiulo
×1
Theodore Papamarkou
×1

Papers

  1. Attention Sinks in Diffusion Language Models
  2. Adaptive Computation Modules: Granular Conditional Computation for Efficient Inference
  3. How to Connect Speech Foundation Models and Large Language Models? What Matters and What Does Not
  4. Task Singular Vectors: Reducing Task Interference in Model Merging
  5. A Simple and Effective L_2 Norm-Based Strategy for KV Cache Compression
  6. Exploiting Activation Sparsity with Dense to Dynamic-k Mixture-of-Experts Conversion
  7. From Latent Graph to Latent Topology Inference: Differentiable Cell Complex Module
  8. Position: Topological Deep Learning is the New Frontier for Relational Learning