LED_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: 1.8994
• Rouge1: 0.3537
• Rouge2: 0.143
• Rougel: 0.297
• Rougelsum: 0.2963
• Gen Len: 20.9033
• Bleu: 0.0675
• Precisions: 0.1559
• Brevity Penalty: 0.6296
• Length Ratio: 0.6837
• Translation Length: 4828.0
• Reference Length: 7062.0
• Precision: 0.8922
• Recall: 0.8771
• F1: 0.8845
• Hashcode: roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)

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: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• 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
• mixed_precision_training: Native AMP

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	5.3039	0.2366	0.0422	0.1815	0.1814	20.3667	0.0181	0.0718	0.6626	0.7084	5003.0	7062.0	0.8757	0.8602	0.8678	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	2.0	38	3.7375	0.2822	0.0849	0.2274	0.2284	20.6433	0.0442	0.1099	0.6381	0.69	4873.0	7062.0	0.8826	0.8679	0.8751	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	3.0	57	3.1526	0.2879	0.088	0.232	0.2322	20.85	0.0444	0.1101	0.6459	0.6958	4914.0	7062.0	0.882	0.8687	0.8752	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	4.0	76	2.7702	0.3011	0.1037	0.2508	0.251	20.8933	0.0508	0.121	0.6366	0.6889	4865.0	7062.0	0.8824	0.87	0.876	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	5.0	95	2.4854	0.3133	0.111	0.2595	0.2597	20.86	0.0514	0.126	0.6305	0.6844	4833.0	7062.0	0.8857	0.8725	0.879	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	6.0	114	2.2629	0.3255	0.1191	0.2732	0.2724	20.9133	0.0578	0.1348	0.6425	0.6933	4896.0	7062.0	0.8872	0.8738	0.8804	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	7.0	133	2.1041	0.3444	0.1375	0.2888	0.2884	20.9133	0.0656	0.1493	0.6279	0.6824	4819.0	7062.0	0.8914	0.8767	0.8839	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	8.0	152	1.9925	0.3617	0.1468	0.3023	0.3022	20.9233	0.0684	0.158	0.6296	0.6837	4828.0	7062.0	0.8925	0.8782	0.8852	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	9.0	171	1.9258	0.3608	0.1447	0.2999	0.2998	20.9133	0.0689	0.1587	0.6286	0.683	4823.0	7062.0	0.8935	0.8785	0.8858	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)
No log	10.0	190	1.8994	0.3537	0.143	0.297	0.2963	20.9033	0.0675	0.1559	0.6296	0.6837	4828.0	7062.0	0.8922	0.8771	0.8845	roberta-large_L17_no-idf_version=0.3.12(hug_trans=4.52.4)

Framework versions

• Transformers 4.52.4
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1