Papers › Compare, Compress and Propagate: Enhancing Neural Architectures with Alignment...
Compare, Compress and Propagate: Enhancing Neural Architectures with Alignment Factorization for Natural Language Inference
Yi Tay, Luu Anh Tuan, Siu Cheung Hui
This paper presents a new deep learning architecture for Natural Language Inference (NLI). Firstly, we introduce a new architecture where alignment pairs are compared, compressed and then propagated to upper layers for enhanced representation learning. Secondly, we adopt factorization layers for efficient and expressive compression of alignment vectors into scalar features, which are then used to augment the base word representations. The design of our approach is aimed to be conceptually simple, compact and yet powerful. We conduct experiments on three popular benchmarks, SNLI, MultiNLI and SciTail, achieving competitive performance on all. A lightweight parameterization of our model also enjoys a ≈3 times reduction in parameter size compared to the existing state-of-the-art models, e.g., ESIM and DIIN, while maintaining competitive performance. Additionally, visual analysis shows that our propagated features are highly interpretable.
In Syntology Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.
Code
No code repository is listed for this paper in the archive or in Syntology's graph.
Code Syntology ran Syntology
Not run by Syntology. Nothing on this page verifies that the listed code works.
Tasks
Results from the paper archive 2025-07-28
| Task | Dataset | Model | Metric | Value | Rank at snapshot | Leaderboard | Report |
|---|---|---|---|---|---|---|---|
| Natural Language Inference | SNLI | 300D CAFE Ensemble | % Test Accuracy | 89.3 | #20 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE Ensemble | % Train Accuracy | 92.5 | #20 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE Ensemble | Parameters | 17.5m | #20 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE | % Test Accuracy | 88.5 | #36 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE | % Train Accuracy | 89.8 | #36 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE | Parameters | 4.7m | #36 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE (no cross-sentence attention) | % Test Accuracy | 85.9 | #65 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE (no cross-sentence attention) | % Train Accuracy | 87.3 | #65 of 98 | Archive leaderboard | report |
| Natural Language Inference | SNLI | 300D CAFE (no cross-sentence attention) | Parameters | 3.7m | #65 of 98 | Archive leaderboard | report |
| Natural Language Inference | SciTail | CAFE | Accuracy | 83.3 | #7 of 13 | 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
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