DICE-BENCH

What is DICE-BENCH?

DICE-BENCH is a benchmark for evaluating large language models’ tool-use capabilities in multi-round, multi-party dialogues, focusing on function selection…

Released

2025-06-28

Evaluates

智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported

Openness

open

Importer

OpenCompass Hub

Reported scores

0

Source provenance

• Original evidence https://hub.opencompass.org.cn/dataset-detail/DICE-BENCH

DICE-BENCH paper, code and dataset

• Paper https://arxiv.org/abs/2506.22853
• Code https://github.com/snuhcc/DICE-Bench
• Dataset https://huggingface.co/datasets/OfficerChul/DICE-BENCH

Which sources cite DICE-BENCH?

• DICE-BENCH

Related 智能体, Agent, 任务执行, Task Execution, 不支持, Unsupported benchmarks

• A-Bench
• A-OKVQA
• AbsPyramid
• ActionAtlas
• Ada-LEval
• AECBench
• AFQMC
• AgentBoard