Share your thoughts, 1 month free Claude Pro on usSee more

Sentence Stress Reasoning on Stress (test)

92.6SSR Accuracy

Human Annotators

Updated 3mo ago

Evaluation Results

Method	Links
Human Annotators 2025.05		92.6	96
gpt-4o 2025.05		86.2	-
StresSLM 2025.05		86.2	-
WhiStress → gpt-4o 2025.05		83.4	-
gpt-4o-mini 2025.05		79.3	-
Gemini-2.5-Pro 2025.05		77.5	-
WhiStress → gpt-4o-mini 2025.05		76.1	-
Llama-3.1-8B-Instruct 2025.05		73.3	-
gpt-4o-audio 2025.05		68.8	-
Qwen2-7B-Instruct 2025.05		67.8	-
WhiStress → Llama-3.1-8B-Instruct 2025.05		65.5	-
WhiStress → Qwen2-7B-Instruct 2025.05		65.5	-
Qwen3-Omni-30B-Instruct 2025.05		64.6	-
Qwen-7B-Chat 2025.05		61.4	-
Audio-Flamingo-3 2025.05		56.8	-
SALMONN 2025.05		55.9	-
Qwen2Audio-7B-Instruct 2025.05		53.2	-
Phi-4-multimodal-instruct 2025.05		52.2	-
WhiStress → Qwen-7B-Chat 2025.05		50.9	-
LLaMA-Omni 2025.05		49.5	-