Examining how large language models (LLMs) perform across various synthetic regression tasks when given (input, output) examples in their context, without any parameter update

Hallucinations (Confabulations) Document-Based Benchmark for RAG. Includes human-verified questions and answers.

A comprehensive guide to LLM evaluation methods designed to assist in identifying the most suitable evaluation techniques for various use cases, promote the adoption of best practices in LLM assessment, and critically assess the effectiveness of these evaluation methods.

A benchmark for prompt injection detection systems.

A collection of LLM related papers, thesis, tools, datasets, courses, open source models, benchmarks

Official implementation for "MJ-Bench: Is Your Multimodal Reward Model Really a Good Judge for Text-to-Image Generation?"

[AAAI 2025] ORQA is a new QA benchmark designed to assess the reasoning capabilities of LLMs in a specialized technical domain of Operations Research. The benchmark evaluates whether LLMs can emulate the knowledge and reasoning skills of OR experts when presented with complex optimization modeling tasks.

LLM-KG-Bench is a Framework and task collection for automated benchmarking of Large Language Models (LLMs) on Knowledge Graph (KG) related tasks.

Benchmark evaluating LLMs on their ability to create and resist disinformation. Includes comprehensive testing across major models (Claude, GPT-4, Gemini, Llama, etc.) with standardized evaluation metrics.

Program synthesis for 3D spatial reasoning

FM-Leaderboard-er allows you to create leaderboard to find the best LLM/prompt for your own business use case based on your data, task, prompts

An app and set of methodologies designed to evaluate the performance of various Large Language Models (LLMs) on the text-to-SQL task. Our goal is to offer a standardized way to measure how well these models can generate SQL queries from natural language descriptions

RTL-Repo: A Benchmark for Evaluating LLMs on Large-Scale RTL Design Projects - IEEE LAD'24

A comprehensive code domain benchmark review of LLM researches.

We introduce a benchmark for testing how well LLMs can find vulnerabilities in cryptographic protocols. By combining LLMs with symbolic reasoning tools like Tamarin, we aim to improve the efficiency and thoroughness of protocol analysis, paving the way for future AI-powered cybersecurity defenses.

The Core AGI Protocol provides a framework to analyze how AGI/ASI might emerge from decentralized, adaptive systems, rather than as the fruit of a single model deployment. It also aims to present orientation as a dynamic and self-evolving Magna Carta, helping to guide the emergence of such phenomena.

Evaluating the Effectiveness of Code-generation Models on Hinglish Prompts

This is a series of Python scripts for zero-shot and chain-of-thought LLM scripting

Python code for the paper "LLMs are zero-shot next-location predictors" by Beneduce et al.

Notable runaway-optimiser-like LLM failure modes on Biologically and Economically aligned AI safety benchmarks for LLM-s with simplified observation format. The benchmark themes include multi-objective homeostasis, (multi-objective) diminishing returns, complementary goods, sustainability, multi-agent resource sharing.