Text Generation
PEFT
Safetensors
English
lora
data-to-text
text-to-data
factual-consistency
hallucination-detection
Instructions to use Loria-MosAIk/xqdt-e2e-gemma3-4b with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- PEFT
How to use Loria-MosAIk/xqdt-e2e-gemma3-4b with PEFT:
from peft import PeftModel from transformers import AutoModelForCausalLM base_model = AutoModelForCausalLM.from_pretrained("google/gemma-3-4b-it") model = PeftModel.from_pretrained(base_model, "Loria-MosAIk/xqdt-e2e-gemma3-4b") - Notebooks
- Google Colab
- Kaggle
File size: 5,808 Bytes
f94cee6 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 | {
"base_model": "google/gemma-3-4b-it",
"cases": [
{
"case": "correct",
"gold": {
"extra": [],
"incorrect": [],
"label": "positive",
"missing": []
},
"input_text": "Blue Spice is a coffee shop in city centre.",
"input_triples": [
"Blue Spice | area | city centre",
"Blue Spice | eat type | coffee shop"
],
"reference_model_output": "All correct",
"reference_parsed_output": {
"extra": [],
"incorrect": [],
"missing": []
},
"sample_id": 0,
"sample_index": 0
},
{
"case": "omitted",
"gold": {
"extra": [],
"incorrect": [],
"label": "negative",
"missing": [
"Mango | buenos aires | Crowne Plaza Hotel",
"The Cricketers | family friendly | yes",
"The Cricketers | eat type | restaurant"
]
},
"input_text": "The coffee shop Blue Spice is based near Crowne Plaza Hotel and has a high customer rating of 5 out of 5.",
"input_triples": [
"Blue Spice | customer rating | 5 out of 5",
"Blue Spice | eat type | coffee shop",
"Blue Spice | near | Crowne Plaza Hotel",
"Mango | buenos aires | Crowne Plaza Hotel",
"The Cricketers | family friendly | yes",
"The Cricketers | eat type | restaurant"
],
"reference_model_output": "| Type | Triple |\n| ---- | ------ |\n| Missing | [S] Mango [P] buenos aires [O] Crowne Plaza Hotel |\n| Missing | [S] The Cricketers [P] eat type [O] restaurant |\n| Missing | [S] The Cricketers [P] family friendly [O] yes |",
"reference_parsed_output": {
"extra": [],
"incorrect": [],
"missing": [
"[S] Mango [P] buenos aires [O] Crowne Plaza Hotel",
"[S] The Cricketers [P] eat type [O] restaurant",
"[S] The Cricketers [P] family friendly [O] yes"
]
},
"sample_id": 2,
"sample_index": 2
},
{
"case": "extra",
"gold": {
"extra": [
"Blue Spice | eat type | pub"
],
"incorrect": [],
"label": "negative",
"missing": []
},
"input_text": "At the riverside, there is a pub called The Blue Spice.",
"input_triples": [
"Blue Spice | area | riverside"
],
"reference_model_output": "| Type | Triple |\n| ---- | ------ |\n| Extra | [S] Blue Spice [P] eat type [O] pub |",
"reference_parsed_output": {
"extra": [
"[S] Blue Spice [P] eat type [O] pub"
],
"incorrect": [],
"missing": []
},
"sample_id": 6,
"sample_index": 6
},
{
"case": "incorrect",
"gold": {
"extra": [],
"incorrect": [
{
"error_type": "wrong_entity",
"incorrect": "Hot | area | riverside",
"original": "Blue Spice | area | riverside",
"replaced_element": "subject"
},
{
"error_type": "wrong_entity",
"incorrect": "Blue Spice | family friendly | Politics",
"original": "Blue Spice | family friendly | no",
"replaced_element": "object"
},
{
"error_type": "wrong_predicate",
"incorrect": "Blue Spice | brand | pub",
"original": "Blue Spice | eat type | pub",
"replaced_element": "predicate"
},
{
"error_type": "wrong_predicate",
"incorrect": "Blue Spice | address | Rainbow Vegetarian Café",
"original": "Blue Spice | near | Rainbow Vegetarian Café",
"replaced_element": "predicate"
},
{
"error_type": "wrong_predicate",
"incorrect": "Blue Spice | class | Chinese",
"original": "Blue Spice | food | Chinese",
"replaced_element": "predicate"
}
],
"label": "negative",
"missing": []
},
"input_text": "Blue Spice pub in riverside serves Chinese food. It is not family friendly and can be found near Rainbow Vegetarian Café.",
"input_triples": [
"Hot | area | riverside",
"Blue Spice | brand | pub",
"Blue Spice | family friendly | Politics",
"Blue Spice | class | Chinese",
"Blue Spice | address | Rainbow Vegetarian Café"
],
"reference_model_output": "| Type | Triple |\n| ---- | ------ |\n| Incorrect | [S] Hot [P] area [O] riverside |\n| Incorrect | [S] Blue Spice [P] brand [O] pub |\n| Incorrect | [S] Blue Spice [P] family friendly [O] Politics |\n| Incorrect | [S] Blue Spice [P] class [O] Chinese |\n| Incorrect | [S] Blue Spice [P] address [O] Rainbow Vegetarian Café |",
"reference_parsed_output": {
"extra": [],
"incorrect": [
"[S] Hot [P] area [O] riverside",
"[S] Blue Spice [P] brand [O] pub",
"[S] Blue Spice [P] family friendly [O] Politics",
"[S] Blue Spice [P] class [O] Chinese",
"[S] Blue Spice [P] address [O] Rainbow Vegetarian Café"
],
"missing": []
},
"sample_id": 13,
"sample_index": 13
}
],
"comparison": "Compare parsed error units after normalization. Byte-identical generation is not required across inference libraries.",
"reference_backend": "ms-swift PtEngine",
"reference_generation": {
"max_tokens": 1024,
"seed": 2023,
"temperature": 0.3
},
"reference_predictions_relative_path": "verifier_train_eval_e2e19/xqdt_outputs/test_predictions_gemma3_4b_6975.json",
"reference_predictions_sha256": "51b9c4e077f7495ca02acbeca9d5798bdbee1ea498c9ec60bf13fd98c08886e0",
"repository": "xqdt-e2e-gemma3-4b",
"schema_version": 1
}
|