Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Instruction Following on MATH benchmark leaderboard · SOTA2 Research

Research Benchmarks

Instruction Following on MATH

49HALT WinsHALT

Updated 5mo ago

Evaluation Results

Method	Links
HALT2025.06		49	59	42	45.4

SOTA Paper

HALT

High Accuracy, Less Talk (HALT): Reliable LLMs through Capability-Aligned Finetuning

Dataset

MATH

Follow for updates

Join SOTA2 Contact

Related Benchmarks

Response correctness and completeness evaluation on Wikipedia Response correctness and completeness evaluation on Math Response correctness and completeness evaluation on Medical Q.Response correctness and completeness evaluation on Coding Instruction Following on appsintro