LLM Deceptiveness and Gullibility Benchmark
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.
Open source
Reactions (1)
Similar products
by description similarity
A multi-source LLM benchmark across text, image, and audio that measures JSON value accuracy per field, not just schema compliance. 7 metrics, full leaderboard.
New LLM outperforming GPT-3.5
news.ycombinator.comRefuel LLM (84.2%) outperforms trained human annotators (80.4%), GPT-3-5-turbo (81.3%), PaLM-2 (82.3%) and Claude (79.3%) across a benchmark of 15 text labeling datasets.
A tool to benchmark LLM APIs (OpenAI
llmapitest.comClaude, local/self-hosted)
LLM Thematic Generalization Benchmark
github.comThematic Generalization Benchmark: measures how effectively various LLMs can infer a narrow or specific "theme" (category/rule) from a small set of examples and anti-examples, then detect which item truly fits that theme among a collection of misleading candidates.
Pencil Puzzle Bench
ppbench.comLLM Benchmark for Multi-Step Verifiable Reasoning