Research map / Architectures and efficiency / RNNs and normalization
End-to-End Differentiable Proving
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Countries S1 | AUC-PR: 90.83 ± 15.4 -> 100.00 ± 0.0 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Countries S2 | AUC-PR: 87.40 ± 11.7 -> 93.04 ± 0.4 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Countries S3 | AUC-PR: 56.68 ± 17.6 -> 77.26 ± 17.0 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Kinship | MRR: 0.60 -> 0.80 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Kinship | HITS@1: 0.48 -> 0.76 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · Nations | MRR: 0.75 -> 0.74 | No clear effect |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · UMLS | MRR: 0.88 -> 0.93 | Helped |
| add ComplEx auxiliary training loss to NTP (ablation) | NTP · UMLS | HITS@1: 0.82 -> 0.87 | Helped |
| use NTPλ instead of ComplEx | NTPλ · Countries S3 | AUC-PR: 48.44 ± 6.3 -> 77.26 ± 17.0 | Helped |
| use NTPλ instead of ComplEx | NTPλ · Kinship | MRR: 0.81 -> 0.80 | Hurt |
| use NTPλ instead of ComplEx | NTPλ · Nations | MRR: 0.75 -> 0.74 | No clear effect |
| use NTPλ instead of ComplEx | NTPλ · UMLS | MRR: 0.89 -> 0.93 | Helped |
Closest papers by meaning
- Learning Reasoning Strategies in End-to-End Differentiable Proving (ICML 2020 · similarity 0.91)
- Differentiable Reasoning on Large Knowledge Bases and Natural Language (AAAI 2020 · similarity 0.89)
- Improving Soft Unification with Knowledge Graph Embedding Methods (ICML 2025 · similarity 0.86)
- Reviving DSP for Advanced Theorem Proving in the Era of Reasoning Models (NeurIPS 2025 · similarity 0.86)
- REFACTOR: Learning to Extract Theorems from Proofs (ICLR 2024 · similarity 0.85)
- ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis (ICML 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.