Vishaal Udandarao's starred repositories
ID-Aligner
Official implement of ID-Aligner
sparo-clip
Separate-head attention read-out
probing-resamplers
Code & data for our NAACL 2024 paper
imgsys-public
imgsys backend
arena-hard-auto
Arena-Hard-Auto: An automatic LLM benchmark.
whatsinthebox
analysis of public NLP corpora
modelcomponents
Decomposing and Editing Predictions by Modeling Model Computation
foundational_fsod
This repository contains the implementation for the paper "Revisiting Few Shot Object Detection with Vision-Language Models"
llama-recipes
Scripts for fine-tuning Meta Llama3 with composable FSDP & PEFT methods to cover single/multi-node GPUs. Supports default & custom datasets for applications such as summarization and Q&A. Supporting a number of candid inference solutions such as HF TGI, VLLM for local or cloud deployment. Demo apps to showcase Meta Llama3 for WhatsApp & Messenger.
gogetcrawl
Extract web archive data using Wayback Machine and Common Crawl
Contamination_Detector
Lightweight tool to identify Data Contamination in LLMs evaluation
better_profanity
Blazingly fast cleaning swear words (and their leetspeak) in strings
GroundingDINO
[ECCV 2024] Official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection"
prepacking
The source code of our work "Prepacking: A Simple Method for Fast Prefilling and Increased Throughput in Large Language Models"
improved-t5
Experiments for efforts to train a new and improved t5
lm-evaluation-harness
A framework for few-shot evaluation of language models.
generating-illustrated-instructions-reproduction
Code for reproducing the paper "Generating Illustrated Instructions."
VisualWebBench
Evaluation framework for paper "VisualWebBench: How Far Have Multimodal LLMs Evolved in Web Page Understanding and Grounding?"