Showing cs.LGShow all
3 papers · 1 filter
cs.LG2026
Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders
Chungpa Lee, Jihoon Kwon, Kyle Min +1
Vision-language models map images and text into a joint embedding space. However, these embeddings often entangle multiple semantic features, which limits their interpretability an…
cs.LG2026
Sink-Token-Aware Pruning for Fine-Grained Video Understanding in Efficient Video LLMs
Kibum Kim, Jiwan Kim, Kyle Min +4
Video Large Language Models (Video LLMs) incur high inference latency due to a large number of visual tokens provided to LLMs. To address this, training-free visual token pruning h…
cs.LG2025
EraseFlow: Learning Concept Erasure Policies via GFlowNet-Driven Alignment
Abhiram Kusumba, Maitreya Patel, Kyle Min +3
Erasing harmful or proprietary concepts from powerful text to image generators is an emerging safety requirement, yet current "concept erasure" techniques either collapse image qua…