lfm2.5-350M-datause-provenance

LoRA SFT of LiquidAI/LFM2.5-350M for data-mention provenance attribute extraction (producer / year / geography / acronym, verbatim from context).

Training

  • base model: LiquidAI/LFM2.5-350M
  • dataset: rafmacalaba/data-use-provenance-sft
  • epochs: 3
  • learning rate: 0.0002
  • LoRA: r=16 alpha=32 dropout=0.05
  • completion-only masking (loss on assistant JSON turn)

Evaluation (holdout, n=3487)

Exact string match of each emitted attribute against the gold label.

attribute tp fp fn precision recall f0.5 f1
producer 1143 296 306 0.7943 0.7888 0.7932 0.7916
year 1085 207 265 0.8398 0.8037 0.8323 0.8213
geography 1715 371 312 0.8221 0.8461 0.8268 0.8339
acronym 1271 212 145 0.8570 0.8976 0.8649 0.8769
overall 5214 1086 1028 0.8276 0.8353 0.8291 0.8314

Verbatim rate (emitted values that are substrings of the context): 6290/6300 = 0.9984

Sample predictions (holdout)

mention target predicted correct verbatim
plant-level data {"geography": "Indonesian"} {"geography": "Indonesian"} geography geography
GIDD database {"geography": "all regions in the world"} {"acronym": "GIDD", "geography": "all regions in the world"} geography geography,acronym
NFHS-3 data {"acronym": "NFHS-3", "geography": "India"} {"acronym": "NFHS-3", "geography": "India"} geography,acronym geography,acronym
data from Hallegatte ( 2012 ) {"year": "2012"} {"geography": "LICs", "year": "2012"} year year,geography
national accounts {"geography": "Turkey"} {"geography": "Turkey"} geography geography
SRF registry of firms {"acronym": "SRF", "geography": "Brazil", "producer": "Brazilian tax authority", "year": "2012"} {"acronym": "SRF", "geography": "Brazil", "producer": "Brazilian tax authority", "year": "2012"} producer,year,geography,acronym producer,year,geography,acronym
Human Capital Index {"geography": "Kenya", "producer": "World Bank", "year": "2020"} {"geography": "Kenya", "producer": "World Bank", "year": "2020"} producer,year,geography producer,year,geography
United Nations population counts by age {"producer": "United Nations"} {"producer": "United Nations"} producer producer
2002 Census {"geography": "Rwanda", "year": "2002"} {"geography": "Rwanda", "year": "2002"} year,geography year,geography
ENCASEH 1997 {"acronym": "ENCASEH", "year": "1997"} {"acronym": "ENCASEH", "year": "1997"} year,acronym year,acronym
Agricultural Wages in India {"acronym": "AWI", "geography": "India", "producer": "Ministry of Agriculture"} {"acronym": "AWI", "geography": "India", "producer": "Ministry of Agriculture"} producer,geography,acronym producer,geography,acronym
COVID-19 Enterprise Survey {"geography": "Chad", "year": "2020"} {"acronym": "ES", "geography": "Chad", "year": "2020"} year,geography year,geography,acronym
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support