Views
No views yet
| Seed | AUC | MRR | NDCG@5 | NDCG@10 |
|---|---|---|---|---|
| 123 | 0.6695 | 0.3221 | 0.3565 | 0.4195 |
| 42 * | 0.6698 | 0.3228 | 0.3590 | 0.4208 |
| mean ± std | 0.6696±0.0001 | 0.3224±0.0004 | 0.3578±0.0012 | 0.4201±0.0006 |
1model:
2 name: glory
3 supported_frameworks:
4 - pytorch
5 architecture:
6 news_encoder:
7 type: mha_pool
8 head_num: 20
9 head_dim: 20
10 attention_hidden_dim: 200
11 graph_encoder:
12 type: gated_graph_conv
13 gnn_num_layers: 3
14 use_graph_type: 0
15 directed: true
16 k_hops: 2
17 num_neighbors: 8
18 entity_neighbors: 10
19 eval_mode: per_impression
20 click_predictor:
21 type: dot_product
22 embedding:
23 size: 300
24 trainable: true
25 dropout_rate: 0.2
26 seed: 42
27 use_entity: true
28 entity_emb_dim: 100
29inputs:
30 title:
31 max_length: 30
32 entity:
33 max_length: 5
34 history:
35 max_length: 50
36 impressions:
37 max_length: 5
38 process_title: true
39 process_abstract: false
40 process_category: true
41 process_subcategory: true
42 process_entities: true
43 process_user_id: false
44training:
45 loss:
46 name: categorical_crossentropy
47 from_logits: true
48 reduction: sum_over_batch_size
49 label_smoothing: 0.0
50 optimizer: adam
51 learning_rate: 0.0002
52 batch_size: 32
53 grad_accum_steps: 1
54 num_epochs: 10
55 gradient_clip_val: 1.0
56 early_stopping:
57 patience: 3
58 min_improvement: 0.01
59 negative_sampling:
60 strategy: random
61 candidates: 4
62evaluation:
63 mode: fast
64 evaluator: default
65 metrics:
66 - auc
67 - mrr
68 - ndcg@5
69 - ndcg@10
70 batch_size: 128newsrex/GLORY-PYTORCH-MIND-small-glove-dev_as_val/
├── model.safetensors ← best seed (42)
├── test_results.json
├── training_run_summary.json
├── seed_123/model.safetensors
├── seed_42/model.safetensors
└── README.md1git clone https://github.com/igor17400/NewsReX.git
2cd NewsReX && uv sync
3
4# Run evaluation with best seed weights
5uv run python src/eval.py \
6 experiment=mind/glove/glory \
7 framework=pytorch \
8 weights=hf://newsrex/GLORY-PYTORCH-MIND-small-glove-dev_as_val/model.safetensors
9
10# Run evaluation with a specific seed
11uv run python src/eval.py \
12 experiment=mind/glove/glory \
13 framework=pytorch \
14 weights=hf://newsrex/GLORY-PYTORCH-MIND-small-glove-dev_as_val/seed_42/model.safetensors1@misc{newsrex2026,
2 title={NewsReX: An Open-Source Multi-Framework for Neural News Recommendation},
3 author={Igor L. R. Azevedo and Toyotaro Suzumura and Yuichiro Yasui},
4 year={2025},
5 eprint={2508.21572},
6 archivePrefix={arXiv},
7 primaryClass={cs.IR},
8 url={https://arxiv.org/abs/2508.21572},
9}