conftrace_

activation reconstruction

5 papers

Explore in graph

Co-occurring keywords · papers mentioning both

gradient-based importance (1) cross model (1) importance-aware compression (1) sparsity objective (1) interpretable (1) sparse (1) memory-efficient fine-tuning (1) network inversion (1) reversible network (1) low-rank compression (1)

Papers

IMPACT: Importance-Aware Activation Space Reconstruction ACL 2026 Scaling and evaluating sparse autoencoders ICLR 2025 Universal Sparse Autoencoders: Interpretable Cross-Model Concept Alignment ICML 2025 Dr2Net: Dynamic Reversible Dual-Residual Networks for Memory-Efficient Finetuning CVPR 2024 How GAN Generators can Invert Networks in Real-Time ACML 2023