LEDv3_ACLsum_all_aspects

floflodebilbao/LEDv3_ACLsum_all_aspects

LEDv3_ACLsum_all_aspects is a machine learning model from floflodebilbao. Use it for the machine learning task on the model card, and read the license before you ship it in a product. It is set up for transformers. The card lists the license as apache-2.0.

should probably proofread and complete it, then remove this comment. --

Downloads · 30 days

10% of all-time downloads

All-time downloads

Public

Parameters

162M

1.3 GB on disk

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors648 MB · 99%

At a glance

Library: transformers
License: apache-2.0
Model type: led
Access: Public
Created: Jul 8, 2025
Updated: Jul 8, 2025
SHA: 7af16275

Base models

allenai/led-base-16384

Model card

From the Hugging Face model README

LEDv3_ACLsum_all_aspects

This model is a fine-tuned version of allenai/led-base-16384 on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 2.2405
Rouge1: 0.3564
Rouge2: 0.1444
Rougel: 0.296
Rougelsum: 0.2951
Gen Len: 20.96
Bleu: 0.0646
Precisions: 0.1586
Brevity Penalty: 0.5945
Length Ratio: 0.6579
Translation Length: 2369.0
Reference Length: 3601.0
Precision: 0.892
Recall: 0.8776
F1: 0.8846
Hashcode: roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 2e-05
train_batch_size: 1
eval_batch_size: 1
seed: 42
gradient_accumulation_steps: 16
total_train_batch_size: 16
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: linear
num_epochs: 10

Training results

Training Loss	Epoch	Step	Validation Loss	Rouge1	Rouge2	Rougel	Rougelsum	Gen Len	Bleu	Precisions	Brevity Penalty	Length Ratio	Translation Length	Reference Length	Precision	Recall	F1	Hashcode
No log	1.0	19	4.0849	0.2835	0.0826	0.2239	0.2243	20.5333	0.0374	0.105	0.6263	0.6812	2453.0	3601.0	0.8837	0.8674	0.8754	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	2.0	38	3.1922	0.2787	0.0853	0.2229	0.2223	20.7667	0.0396	0.1061	0.6161	0.6737	2426.0	3601.0	0.8767	0.8646	0.8705	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	3.0	57	2.8716	0.292	0.098	0.2328	0.2329	20.84	0.0461	0.1181	0.6082	0.6679	2405.0	3601.0	0.8812	0.8688	0.8749	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	4.0	76	2.6529	0.3166	0.1192	0.2573	0.2566	20.92	0.0576	0.1366	0.6104	0.6695	2411.0	3601.0	0.8861	0.8725	0.8792	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	5.0	95	2.5101	0.3441	0.1353	0.282	0.2813	20.94	0.0613	0.1495	0.593	0.6568	2365.0	3601.0	0.8895	0.8763	0.8828	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	6.0	114	2.3985	0.3501	0.1415	0.2909	0.2912	20.92	0.0616	0.1514	0.5983	0.6606	2379.0	3601.0	0.8913	0.8771	0.884	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	7.0	133	2.3215	0.3557	0.1398	0.295	0.2949	20.9667	0.0608	0.1545	0.593	0.6568	2365.0	3601.0	0.8919	0.8775	0.8846	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	8.0	152	2.2783	0.3494	0.1417	0.2922	0.2918	20.9333	0.0637	0.1561	0.588	0.6532	2352.0	3601.0	0.8907	0.8769	0.8837	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	9.0	171	2.2467	0.3566	0.145	0.297	0.2968	20.96	0.0649	0.1591	0.5926	0.6565	2364.0	3601.0	0.8921	0.8775	0.8846	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)
No log	10.0	190	2.2405	0.3564	0.1444	0.296	0.2951	20.96	0.0646	0.1586	0.5945	0.6579	2369.0	3601.0	0.892	0.8776	0.8846	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.53.1)

Framework versions

Transformers 4.53.1
Pytorch 2.7.0+cu126
Datasets 3.6.0
Tokenizers 0.21.1

Library: transformers
Type: led
License: apache-2.0
Created: Jul 8, 2025
Updated: Jul 8, 2025