[EMNLP 2024 & AAAI 2026] A powerful toolkit for compressing large models including LLMs, VLMs, and video generative models.

This is a collection of our research on efficient AI, covering hardware-aware NAS and model compression.

[NeurIPS 2025] HoliTom: Holistic Token Merging for Fast Video Large Language Models

[CVPR 2025] DivPrune: Diversity-based Visual Token Pruning for Large Multimodal Models

[CVPR 2025] PACT: Pruning and Clustering-Based Token Reduction for Faster Visual Language Models

Official implementation of CVPR 2024 paper "vid-TLDR: Training Free Token merging for Light-weight Video Transformer".

Official Implementation (Pytorch) of the "Representation Shift: Unifying Token Compression with FlashAttention", ICCV 2025

A token pruning method that accelerates ViTs for various tasks while maintaining high performance.

Implementation of ICCV 2025 paper "Growing a Twig to Accelerate Large Vision-Language Models".

An implementation of LazyLLM token pruning for LLaMa 2 model family.

Boundary-discovery and anti-self-deception framework for AI efficiency research. Produces falsifiable, condition-specific verdicts. First validated result: a hard failure boundary for token pruning.

Token pruning and sparse attention for efficient Vision-Language Model inference

Visual token budget simulator for efficient VLM inference