TIME-Module: Mapping — flan-t5-small
Model Description
Temporal span prediction model that maps natural language time expressions to structured temporal spans. Given a query with temporal context (reference time, timezone), predicts ISO-format start/end timestamps. Best performing model on the quality-filtered dataset.
Training Details
- Base Model: google/flan-t5-small
- Architecture: T5ForConditionalGeneration
- Dataset: Pieces/temporal-span-prediction-v-0-3-1-quality (41,927 train / ~5K val / ~5K test)
- Training Steps: 90,000
- Learning Rate: 3e-4
- Batch Size: 16 × 2
- Mixed Precision: bf16
- Hardware: NVIDIA RTX 4090 (24 GB)
Results
| Metric | Value |
|---|---|
| exact_match | 56.36% |
| tmr | ~59.5% |
| avg_iou | ~70.3% |
| parse_success | 99.9% |
Usage
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
tokenizer = AutoTokenizer.from_pretrained('Pieces/time-mapping-flan-t5-small-quality-best')
model = AutoModelForSeq2SeqLM.from_pretrained('Pieces/time-mapping-flan-t5-small-quality-best')
inputs = tokenizer('Map: tomorrow at 3pm | ref_time: 2026-01-15T10:00:00 | tz: UTC', return_tensors='pt')
outputs = model.generate(**inputs)
result = tokenizer.decode(outputs[0], skip_special_tokens=True)
Part of the TIME-Module Project
This model is part of the TIME (Temporal Intent, Mapping, and Extraction) module, a suite of models for understanding and processing temporal information in natural language.
Related models:
- Pieces/time-classification-flan-t5-small-split-best — Intent classification
- Pieces/time-mapping-flan-t5-small-quality-best — Span prediction (best)
- Pieces/time-mapping-t5gemma-270m-best — Span prediction (T5Gemma)
Citation
@software{time_module,
title={TIME-Module: Temporal Intent, Mapping, and Extraction},
author={Pieces},
year={2026},
url={https://huggingface.co/Pieces}
}