Papers › BERT got a Date: Introducing Transformers to Temporal Tagging

BERT got a Date: Introducing Transformers to Temporal Tagging

30 Sep 2021arXiv:2109.14927archive 2025-07-28

Satya Almasian, Dennis Aumiller, Michael Gertz

Temporal expressions in text play a significant role in language understanding and correctly identifying them is fundamental to various retrieval and natural language processing systems. Previous works have slowly shifted from rule-based to neural architectures, capable of tagging expressions with higher accuracy. However, neural models can not yet distinguish between different expression types at the same level as their rule-based counterparts. In this work, we aim to identify the most suitable transformer architecture for joint temporal tagging and type classification, as well as, investigating the effect of semi-supervised training on the performance of these systems. Based on our study of token classification variants and encoder-decoder architectures, we present a transformer encoder-decoder model using the RoBERTa language model as our best performing system. By supplementing training resources with weakly labeled data from rule-based systems, our model surpasses previous works in temporal tagging and type classification, especially on rare classes. Our code and pre-trained experiments are available at: https://github.com/satya77/Transformer_Temporal_Tagger

PaperPDFCode

In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

Code

satya77/Transformer_Temporal_Tagger officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

ClassificationDecoderLanguage ModelingLanguage ModellingRetrievalTemporal TaggingToken Classificationtoken-classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Temporal Tagging TempEval-3 R2R Strict Detection (Pr.) 96.37 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Strict Detection (Re.) 96.37 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Relaxed Detection (F1) 100 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Relaxed Detection (Pr.) 100 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Relaxed Detection (Re.) 100 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Strict Detection (F1) 96.37 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 R2R Type 90.43 #1 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Strict Detection (Pr.) 94.11 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Strict Detection (Re.) 81.01 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Relaxed Detection (F1) 92.52 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Relaxed Detection (Pr.) 100 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Relaxed Detection (Re.) 86.09 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Strict Detection (F1) 87.07 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 B2B Type 83.79 #2 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Strict Detection (Pr.) 82.72 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Strict Detection (Re.) 85.79 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Relaxed Detection (F1) 92.60 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Relaxed Detection (Pr.) 90.95 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Relaxed Detection (Re.) 94.35 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Strict Detection (F1) 84.21 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 DateBERT Type 86.21 #3 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Strict Detection (Pr.) 81.83 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Strict Detection (Re.) 79.56 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Relaxed Detection (F1) 90.08 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Relaxed Detection (Pr.) 91.37 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Relaxed Detection (Re.) 88.84 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Strict Detection (F1) 80.67 #4 of 5 Archive leaderboard report
Temporal Tagging TempEval-3 BERT-base Type 82.00 #4 of 5 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Methods

AdamAttentionAttention DropoutBERTDense ConnectionsDropoutLayer NormalizationLinear LayerLinear Warmup With Linear DecayMulti-Head AttentionResidual ConnectionRoBERTaSoftmaxWeight DecayWordPiece

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections