Meta Cognitive Calibration Benchmark for Large Language Models

A benchmark dataset from Kaggle for evaluating the cognitive calibration of large language models. The dataset likely contains tasks or prompts designed to measure how well an LLM's confidence aligns with its accuracy. Its specific size, structure, and creation details require verification after download.

Use Cases

• Benchmarking LLM confidence calibration on reasoning tasks (inferred from domain, verify after download)
• Analyzing the relationship between model uncertainty and answer correctness (inferred from domain, verify after download)
• Training or fine-tuning models for improved self-assessment capabilities (inferred from domain, verify after download)

Strengths

• Published on Kaggle, a major platform for data science resources.
• Platform tags indicate a focus on LLM benchmarking and cognitive science.

Limitations

• Metadata is minimal; actual content requires verification after download.
• Row count, column definitions, and file formats are unknown.
• License, author, and last update information are unavailable.

Provenance

Source

Kaggle

Collection Method

Unknown

Time Range

Unknown

Freshness

Last update date is unknown; freshness unverified.

Geography

Unknown