llms-evalution

A set of scripts to generate full attention-head heatmaps for transformer-based LLMs

Repository for the paper: Revisiting LLM Evaluation through Mechanism Interpretability: a New Metric and Model Utility Law

Biological Information Extraction from Large Language Models (LLMs) (Journal of Computational Biology 2025)

Measure of estimated confidence for non-hallucinative nature of outputs generated by Transformer-based Language Models.

Benchmarking LLMs (large language models) on leetcode algorithmic challenges

250 LLM Benchmarks & Evaluation Datasets

A set of scripts to generate full attention-head heatmaps for transformer-based LLMs

BAREC-ST-2026. The Second Shared Task on Sentence-level Readability Assessment (Strict Track): An NLP competition on fine-grained Arabic readability classification across 19 levels.

Open framework for building AI-readable websites that improve discoverability across modern AI-powered search experiences.

A lightweight black-box agent evaluator using YAML specifications to score task completion.

Preliminary Results: Biological Information Extraction from LLMs (BioNLP 2023)

An interactive magazine-style exploration of the GSM-Symbolic research paper on LLM reasoning limitations.

Applied and evaluated bias mitigation strategies on modern LLMs.

AI Evals for the regression models

Hate Data entry in Calendars? Provide URL or Text information related to events, meetings and visits.Collect all the interesting events (URLS or texts) in a dedicated google calendar to see these in a central place. Move the most interesting event to your main calendar with one click in calendar

Transformer-based multilingual and cross-lingual abstractive text summarization for Marathi and English with systematic evaluation.

Enhancing Fault Isolation in Hardware Systems Using Large Language Models (LLMs). Master's Degree Thesis developed in 2025 during an internship at MBDA.

Framework for benchmarking and analyzing LLMs on simple games