eval-harness

reference

gold/summeval_100 gold

Promoted as gold: the yardstick a decision rests on.
Promoted from person the dataset's authors, 2026-10-15: promoted before labels existed (0.6.0): its tier was given when it was made

01what it is

tier
gold
dataset
summeval_100
promoted from
person the dataset's authors
why
promoted before labels existed (0.6.0): its tier was given when it was made
items
100
created
2026-10-15T13:03:29Z

02scored against it · 41

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

armmetricmeanitems
se_gemma_l_n100_pairwise_v1pair_agreement0.825100inspect
se_qwen_s_n100_pairwise_v1pair_agreement0.812100inspect
se_random_n100_pairwise_v1pair_agreement0.473100inspect
se_gemma_l_pubtext_n100_v1 · pairtau0.551100inspect
se_openai_l_n100_v1tau0.540100inspect
se_gemma_l_pubtextcw_n100_v1 · pairtau0.540100inspect
se_gemma_l_n100_v1tau0.538100inspect
se_anthropic_l_n100_v1tau0.528100inspect
se_openai_m_n100_v1tau0.528100inspect
se_qwen37max_n100_v1tau0.520100inspect
se_qwen_s_n100_v1tau0.519100inspect
se_glm_m_n100_v1tau0.517100inspect
se_qwen_m_n100_v1tau0.507100inspect
se_glm_l_n100_v1tau0.492100inspect
se_deepseek_m_n100_v1tau0.486100inspect
se_anthropic_s_n100_v1tau0.484100inspect
se_anthropic_m_n100_v1tau0.480100inspect
se_gemma_m_n100_v1tau0.477100inspect
se_deepseek_l_n100_v1tau0.469100inspect
se_mistral_m_n100_v1tau0.468100inspect
se_deepseek_s_n100_v1tau0.459100inspect
se_mistral_l_n100_v1tau0.455100inspect
se_llama_l_n100_v1tau0.454100inspect
se_glm_s_n100_v1tau0.446100inspect
se_mistral_s_n100_v1tau0.427100inspect
se_gemma_s_n100_v1tau0.420100inspect
se_openai_s_n100_v1tau0.411100inspect
se_llama_m_n100_v1tau0.406100inspect
se_unievalpub_n100_v1tau0.372100inspect
se_llama_s_n100_v1tau0.353100inspect
se_unieval_n100_v1tau0.348100inspect
se_bartscore_n100_v1tau0.256100inspect
se_rouge2_n100_v1tau0.158100inspect
se_rouge1_n100_v1tau0.149100inspect
se_rougeL_n100_v1tau0.143100inspect
se_coverage_n100_v1tau0.101100inspect
se_length_n100_v1tau0.039100inspect
se_random_n100_v1tau0.003100inspect
se_gemma_l_n100_pick_v1top10.364100inspect
se_qwen_s_n100_pick_v1top10.263100inspect
se_random_n100_pick_v1top10.070100inspect