LLM-Drop

university

https://github.com/CASE-Lab-UMD/LLM-Drop

AI & ML interests

Efficient and adaptive foundation models across language and multimodal intelligence.

Recent Activity

shwai-he submitted a paper about 21 hours ago

Disentangling Representation Evolution in Transformers through Directional Decomposition

shwai-he updated a collection about 2 months ago

shwai-he updated a collection about 2 months ago

View all activity

Papers

Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?

Demystifying When Pruning Works via Representation Hierarchies

View all Papers

submitted a paper to Daily Papers about 21 hours ago

Disentangling Representation Evolution in Transformers through Directional Decomposition

Paper • 2609.15975 • Published 3 days ago • 4

updated a collection about 2 months ago

Efficient_UG

Resources for “Understanding and Harnessing Sparsity in Unified Multimodal Models”, including the paper and efficient BAGEL-MoE checkpoints. • 3 items • Updated Jul 31 • 1

updated a collection about 2 months ago

VLADrop

Resources for Drop-Then-Recovery (DTR), including VLADrop checkpoints for studying and recovering redundancy in vision-language-action models. • 1 item • Updated Jul 31 • 1

updated 2 collections about 2 months ago

EffiR

Resources for paper “Making Large Language Models Efficient Dense Retrievers” (ACL 2026), including the paper and efficient Mistral-based dense retrie • 5 items • Updated Jul 31 • 1

LLM-Drop

Resources for studies on redundancy in LLMs, including layer dropping and representation-hierarchy-based pruning. • 16 items • Updated Jul 31 • 7

updated a Space about 2 months ago

README

updated a collection 3 months ago

LLM-Drop

Resources for studies on redundancy in LLMs, including layer dropping and representation-hierarchy-based pruning. • 16 items • Updated Jul 31 • 7

submitted a paper to Daily Papers 3 months ago

Drop-Then-Recovery: How Redundant Are Vision-Language-Action Models?

Paper • 2606.27755 • Published Jun 26 • 6

authored 2 papers 5 months ago

Selective Reflection-Tuning: Student-Selected Data Recycling for LLM Instruction-Tuning

Paper • 2402.10110 • Published Feb 15, 2024 • 3

What Matters in Transformers? Not All Attention is Needed

Paper • 2406.15786 • Published Jun 22, 2024 • 33