sonar-sae / data_night3 /validation_report.json
nickypro's picture
Upload data_night3/validation_report.json with huggingface_hub
8fa85de verified
Raw
History Blame Contribute Delete
11.8 kB
{
"token_position_grid": {
"n": 30,
"sbert_cos_mean": 0.9094,
"sbert_cos_min": 0.8466,
"token_f1_mean": 0.7263,
"token_f1_min": 0.5854,
"exact_match_rate": 0.0,
"examples": [
{
"label": "marsh@p8/f2",
"orig": "Bridge, cliff, curtain, drawer, engine, fountain, helmet, marsh, kitchen, meadow, pillow, pocket, shovel, statue, trumpet, valley were read aloud twice.",
"decoded": "Bridge, crane, slope, drawer, machine, fountain, helmet, swamp, kitchen, lawn, mat, pocket, pillow, statue, trumpet, valley were read loudly two times.",
"token_f1": 0.585,
"sbert_cos": 0.847
},
{
"label": "mushroom@p9/f2",
"orig": "Bridge, cliff, curtain, drawer, engine, fountain, helmet, jacket, mushroom, meadow, pillow, pocket, shovel, statue, trumpet, valley were read aloud twice.",
"decoded": "Bridge, cliff, curtain, drawer, engine, fountain, helmet, jacket, mosquito, grass, pillow, pocket, pillow, statue, trumpet, valley were read loudly two times.",
"token_f1": 0.732,
"sbert_cos": 0.871
},
{
"label": "think@p14/f0",
"orig": "Table, river, candle, mirror, garden, pencil, button, ladder, basket, window, carpet, lantern, ribbon, think, kettle, saddle were on the list.",
"decoded": "Table, river, lamp, mirror, garden, pencil, button, scale, basket, window, carpet, lantern, ribbon, thought, kettle, saddle, was on the list.",
"token_f1": 0.8,
"sbert_cos": 0.972
},
{
"label": "marsh@p8/f2",
"orig": "Bridge, cliff, curtain, drawer, engine, fountain, helmet, marsh, kitchen, meadow, pillow, pocket, shovel, statue, trumpet, valley were read aloud twice.",
"decoded": "Bridge, crane, slope, drawer, machine, fountain, helmet, swamp, kitchen, lawn, mat, pocket, pillow, statue, trumpet, valley were read loudly two times.",
"token_f1": 0.585,
"sbert_cos": 0.847
},
{
"label": "jungle@p3/f1",
"orig": "Anchor, barrel, jungle, feather, island, lemon, magnet, needle, orchard, pebble, quilt, raft, sparrow, tunnel, violin, wagon appeared in that order.",
"decoded": "Anchor, barrel, vine, jungle, leaf, jar, magnet, needle, jewelry, plank, woodpecker, pitch, funnel, tunnel, violin, wagon appeared in that order.",
"token_f1": 0.6,
"sbert_cos": 0.86
}
]
},
"spec_edit_corpus_base": {
"n": 30,
"sbert_cos_mean": 0.9817,
"sbert_cos_min": 0.87,
"token_f1_mean": 0.9445,
"token_f1_min": 0.7619,
"exact_match_rate": 0.4667,
"examples": [
{
"label": "word-swap",
"orig": "Emily's eyes scanned the yellowed pages of a tattered book, her fingers tracing the faint lines of text as she searched for a specific passage.",
"decoded": "Emily's eyes sifted the yellowish pages of a rotted book, her fingers traced the faint lines of text as she searched for a particular passage.",
"token_f1": 0.8,
"sbert_cos": 0.94
},
{
"label": "word-swap",
"orig": "In the end, it was a combination of old-fashioned detective work and innovative thinking that led to the eventual breakthrough in the case.",
"decoded": "In the end, it was a combination of old-fashioned detective work and innovative thinking that led to the eventual breakthrough in the case.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "word-swap",
"orig": "Irving's account portrayed the Dutch as warm, hospitable, and Christmas-loving, emphasizing their role in shaping the holiday traditions of New Amsterdam (prese",
"decoded": "Irving's account portrayed the Dutch as warm, hospitable, and Christmas-loving, emphasizing their role in shaping the holiday traditions of New Amsterdam (now N",
"token_f1": 0.943,
"sbert_cos": 0.999
},
{
"label": "delete-clause",
"orig": "The Fund, which holds a royalty interest in certain of these restaurants, benefits from the parent company's growth and profitability.",
"decoded": "The Fund, which has a royalty interest in some of these restaurants, benefits from the growth and profitability of the mother company.",
"token_f1": 0.762,
"sbert_cos": 0.939
},
{
"label": "register",
"orig": "Our volunteers are the backbone of our organisation, and we're grateful for their dedication and commitment.",
"decoded": "Our volunteers are the backbone of our organization, and we are thankful for their dedication and commitment.",
"token_f1": 0.788,
"sbert_cos": 0.964
}
]
},
"spec_edit_corpus_gold": {
"n": 30,
"sbert_cos_mean": 0.9821,
"sbert_cos_min": 0.8805,
"token_f1_mean": 0.9562,
"token_f1_min": 0.8,
"exact_match_rate": 0.4333,
"examples": [
{
"label": "word-swap",
"orig": "Emily's eyes scanned the yellowed pages of a tattered letter, her fingers tracing the faint lines of text as she searched for a specific passage.",
"decoded": "Emily's eyes sifted the yellowed pages of a broken letter, her fingers traced the faint lines of text as she searched for a particular passage.",
"token_f1": 0.84,
"sbert_cos": 0.923
},
{
"label": "word-swap",
"orig": "In the end, it was a combination of new-fashioned detective work and innovative thinking that led to the eventual breakthrough in the case.",
"decoded": "In the end, it was a combination of new-style detective work and innovative thinking that led to the eventual breakthrough in the case.",
"token_f1": 0.958,
"sbert_cos": 0.99
},
{
"label": "word-swap",
"orig": "Irving's account portrayed the Dutch as warm, hospitable, and Christmas-loving, emphasizing their role in shaping the holiday traditions of New Amsterdam (prese",
"decoded": "Irving's account portrayed the Dutch as warm, hospitable, and Christmas-loving, emphasizing their role in shaping the holiday traditions of New Amsterdam (now N",
"token_f1": 0.943,
"sbert_cos": 0.994
},
{
"label": "negation",
"orig": "With some salary cap considerations in mind, the Saints will not be monitoring the early starter pieces and any bold draft moves to get added cushion.",
"decoded": "With some salary cap considerations in mind, The Saints won't be monitoring the early starters and any daring draft moves to get extra cushion.",
"token_f1": 0.8,
"sbert_cos": 0.959
},
{
"label": "word-swap",
"orig": "Emily's eyes scanned the yellowed pages of a tattered letter, her fingers tracing the faint lines of text as she searched for a specific passage.",
"decoded": "Emily's eyes sifted the yellowed pages of a broken letter, her fingers traced the faint lines of text as she searched for a particular passage.",
"token_f1": 0.84,
"sbert_cos": 0.923
}
]
},
"paraphrase_bank_a": {
"n": 15,
"sbert_cos_mean": 0.968,
"sbert_cos_min": 0.8289,
"token_f1_mean": 0.9486,
"token_f1_min": 0.8571,
"exact_match_rate": 0.6,
"examples": [
{
"label": "mined_orbit",
"orig": "As the world continues to evolve, policymakers will need to navigate the challenges and opportunities presented by future trade agreements.",
"decoded": "As the world continues to evolve, policymakers will need to navigate the challenges and opportunities presented by future trade agreements.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "template_active_passive",
"orig": "The fuzzy ladder scoured the shallow hand.",
"decoded": "The fuzzy ladder swept the shallow hand.",
"token_f1": 0.857,
"sbert_cos": 0.891
},
{
"label": "template_cleft_passive",
"orig": "It was the sticky waiter that stacked the full number.",
"decoded": "It was the sticky waiter who stacked the full number.",
"token_f1": 0.9,
"sbert_cos": 0.995
},
{
"label": "template_active_passive",
"orig": "The fuzzy ladder scoured the shallow hand.",
"decoded": "The fuzzy ladder swept the shallow hand.",
"token_f1": 0.857,
"sbert_cos": 0.891
},
{
"label": "template_active_passive",
"orig": "The hairy jungle hauled the velvet violin.",
"decoded": "The hairy jungle dragged the velvet violin.",
"token_f1": 0.857,
"sbert_cos": 0.943
}
]
},
"paraphrase_bank_b": {
"n": 15,
"sbert_cos_mean": 0.9631,
"sbert_cos_min": 0.8221,
"token_f1_mean": 0.9475,
"token_f1_min": 0.8387,
"exact_match_rate": 0.5333,
"examples": [
{
"label": "mined_orbit",
"orig": "As the world progresses, policymakers will have to address the challenges and opportunities that future trade agreements present.",
"decoded": "As the world progresses, policymakers will have to address the challenges and opportunities that future trade agreements present.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "template_active_passive",
"orig": "The shallow hand was scoured by the fuzzy ladder.",
"decoded": "The shallow hand was scrubbed by the fuzzy ladder.",
"token_f1": 0.889,
"sbert_cos": 0.886
},
{
"label": "template_cleft_passive",
"orig": "The full number was stacked by the sticky waiter.",
"decoded": "The full number was stacked by the sticky waiter.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "mined_orbit",
"orig": "For further information about the conference, including registration and the schedule, kindly visit the [NEA website].",
"decoded": "For further information about the conference, including registration and the schedule, please visit [NEA's website].",
"token_f1": 0.839,
"sbert_cos": 0.982
},
{
"label": "template_adverb_move",
"orig": "Steadily, the slippery cabin packed the room.",
"decoded": "Steadily, the slippery cabin filled the room.",
"token_f1": 0.857,
"sbert_cos": 0.926
}
]
},
"construction_specs": {
"n": 30,
"sbert_cos_mean": 0.9052,
"sbert_cos_min": 0.6537,
"token_f1_mean": 0.8748,
"token_f1_min": 0.6667,
"exact_match_rate": 0.2667,
"examples": [
{
"label": "L04_002",
"orig": "The quarry laughed noisily.",
"decoded": "The quarry laughed loudly.",
"token_f1": 0.75,
"sbert_cos": 0.806
},
{
"label": "L04_039",
"orig": "The mattress smiled hard.",
"decoded": "The mattress smiled hard.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "L04_042",
"orig": "The year relaxed hard.",
"decoded": "The year relaxed hard.",
"token_f1": 1.0,
"sbert_cos": 1.0
},
{
"label": "L06_008",
"orig": "The clever dell plastered the donkey.",
"decoded": "The smart dell gelled the donkey.",
"token_f1": 0.667,
"sbert_cos": 0.839
},
{
"label": "L12_043",
"orig": "The plump jungle followed the rotten tine abroad near the dank pavement.",
"decoded": "The thick jungle followed the rotten tin to abroad near the wet pavement.",
"token_f1": 0.72,
"sbert_cos": 0.882
}
],
"content_word_recall_mean": 0.8208
}
}