Latest Articles for Benchmark

A new benchmark for machine unlearning enhances evaluation and comparison of methods.

A new method improves code generation accuracy using external documents.

CEBench helps businesses and researchers assess LLMs while managing costs and performance.

Research highlights in-context learning abilities in large language models.

New framework evaluates SLAM performance under challenging conditions.

New benchmark assesses how video-language models handle inaccuracies effectively.

Evaluating how LLMs create persuasive text across various topics.

This study benchmarks Language Models' performance using Italian INVALSI tests.

A benchmark tool advances active learning strategies in machine learning.

This article assesses the effectiveness of large language models in creating hardware assertions.

A new AI system enhances accessibility for users with visual impairments through better screen reading.

A benchmark of minimal pairs aims to improve understanding of Russian grammar by language models.

A new model streamlines data analysis in vast datasets using sketches.

A new benchmark for improving biophysical sequence optimization methods.

This study presents a fresh method for detecting anomalies in various contexts.

New benchmark improves evaluation of multimodal models by minimizing biases.

New benchmark aids in predicting enzyme behavior using machine learning.

New models produce high-quality video descriptions effectively.

A comprehensive benchmark enhances evaluation of vision-language models for biological image analysis.

A new benchmark for assessing large language models in hypothesis testing.

A new benchmark addresses challenges in code retrieval for developers.

This research examines how visual issues impact Visual Question Answering models.

NFARD offers innovative methods to protect deep learning model copyrights.

A new model improves safety monitoring for large language models against harmful content.

A look into how Bayesian optimization addresses high-dimensional challenges.

A new method to assess data analytics agents for better business insights.

Introducing MaxCut-Bench for consistent algorithm assessment in optimization challenges.

Improving how models handle evidence in long documents builds user trust.

Assessing LLM capabilities using grid-based games like Tic-Tac-Toe and Connect Four.

A new benchmark aims to assess AI safety risks effectively.

Combining visuals and language enhances hardware code generation accuracy.

A new benchmark addresses the need for standard evaluation in spatio-temporal prediction.

New methods improve testing for language models, focusing on key performance areas.

A novel benchmark to evaluate graph learning methods tackling heterophily and heterogeneity.

A framework to assess LLMs' abilities in data-related tasks with code interpreters.

A look into how CLIP processes negation in language.

Establishing a benchmark to evaluate fairness in graph learning methods.

Exploring how language models tackle reasoning tasks effectively.

A new benchmark assesses language models on scientific coding challenges across multiple fields.

A new model improves how machines read charts, even without labels.