DrishtiShrrrma

Unlearning_LLM

[ACL 2024] Code and data for "Machine Unlearning of Pre-trained Large Language Models"

Multi-Agent_Pickup_and_Delivery

Implementations of various algorithms used to solve the problem of Multi-Agent Pickup and Delivery (a generalization of Multi-Agent Path Finding).

Qwen2-VL

Qwen2-VL is the multimodal large language model series developed by Qwen team, Alibaba Cloud.

WorkBench

WorkBench: a Benchmark Dataset for Agents in a Realistic Workplace Setting.

LlavaGuard

distilling-step-by-step

reward-bench

RewardBench: the first evaluation tool for reward models.

L1B3RT45

JAILBREAK PROMPTS FOR LIBERATING AI MODELS

lm-evaluation-harness

A framework for few-shot evaluation of language models.

maya-dataset-creation

The Repository contains the code for dataset creation for the Training the Maya: Multilingual Aya Model

mlmm-evaluation

Multilingual Large Language Models Evaluation Benchmark

LLaVA

[NeurIPS'23 Oral] Visual Instruction Tuning (LLaVA) built towards GPT-4V level capabilities and beyond.

DOSA

Dataset of of Social Artifacts from Different Indian Geographical Subcultures

culture-llm

nllb-serve

Meta's "No Language Left Behind" models served as web app and REST API

M4GT-Bench

NAACL-2024-SemEval-TASK-8C

Code for the paper : Black-Box Word-Level Text Boundary Detection in Partially Machine Generated Texts

aya_rm_multilingual

Repository for Aya Expedition Project : Reward Model Multilingual

GoldenFace

An Image Processing Library About Calculating Face Golden Ratio, Facial Cosine Similarity and More

face_rating

Face/Beauty Rating with both the traditional ML approaches and Convolutional Neural Network Approach

Multi-LLM-Agent

Multi-Agents-Debate

MAD: The first work to explore Multi-Agent Debate with Large Language Models :D

llm-compression-benchmark

LLM Compression Benchmark

massive

Tools and Modeling Code for the MASSIVE dataset

mgt-detection-benchmark

Multilingual machine-generated text detection benchmark

llama-cpp-perplexity-scorecard

Run llama.cpp perplexity test and save results to a cloud datastore for analysis and comparison

instant-ngp

Instant neural graphics primitives: lightning fast NeRF and more

interpretability-starter

🧠 Starter templates for doing interpretability research

PROMST

Automatic prompt optimization framework for multi-step agent tasks.

promptfoo

Test your prompts, agents, and RAGs. Red teaming, pentesting, and vulnerability scanning for LLMs. Compare performance of GPT, Claude, Gemini, Llama, and more. Simple declarative configs with command line and CI/CD integration.