Views
No views yet
| Seed | AUC | MRR | NDCG@5 | NDCG@10 |
|---|---|---|---|---|
| 123 | 0.6512 | 0.3023 | 0.3342 | 0.3966 |
| 42 * | 0.6520 | 0.3031 | 0.3335 | 0.3980 |
| 456 | 0.6456 | 0.3017 | 0.3324 | 0.3958 |
| mean ± std | 0.6496±0.0028 | 0.3024±0.0006 | 0.3334±0.0007 | 0.3968±0.0009 |
1model:
2 name: naml
3 architecture:
4 news_encoder:
5 type: cnn_multi_view
6 views:
7 title:
8 type: cnn
9 filter_num: 400
10 kernel_size: 3
11 activation: relu
12 attention_query_dim: 200
13 abstract:
14 type: cnn
15 filter_num: 400
16 kernel_size: 3
17 activation: relu
18 attention_query_dim: 200
19 category:
20 type: embedding
21 embedding_dim: 100
22 subcategory:
23 type: embedding
24 embedding_dim: 100
25 view_attention_query_dim: 200
26 user_encoder:
27 type: additive_attention
28 attention_query_dim: 200
29 num_attention_heads: 10
30 click_predictor:
31 type: dot_product
32 embedding:
33 size: 300
34 trainable: true
35 dropout_rate: 0.2
36 seed: 42
37inputs:
38 title:
39 max_length: 32
40 abstract:
41 max_length: 50
42 history:
43 max_length: 50
44 impressions:
45 max_length: 5
46 process_title: true
47 process_abstract: true
48 process_category: true
49 process_subcategory: true
50 process_user_id: false
51training:
52 loss:
53 name: categorical_crossentropy
54 from_logits: true
55 reduction: sum_over_batch_size
56 label_smoothing: 0.0
57 optimizer: adam
58 learning_rate: 0.0001
59 batch_size: 256
60 num_epochs: 10
61 early_stopping:
62 patience: 3
63 min_improvement: 0.01
64 negative_sampling:
65 strategy: random
66 candidates: 4
67evaluation:
68 mode: fast
69 metrics:
70 - auc
71 - mrr
72 - ndcg@5
73 - ndcg@10
74 batch_size: 512newsrex/NAML-JAX-MIND-small-glove-random/
├── model.safetensors ← best seed (42)
├── test_results.json
├── training_run_summary.json
├── seed_123/model.safetensors
├── seed_42/model.safetensors
├── seed_456/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/naml \
7 framework=jax \
8 weights=hf://newsrex/NAML-JAX-MIND-small-glove-random/model.safetensors
9
10# Run evaluation with a specific seed
11uv run python src/eval.py \
12 experiment=mind/glove/naml \
13 framework=jax \
14 weights=hf://newsrex/NAML-JAX-MIND-small-glove-random/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}