Research map / Learning theory and algorithms / Prediction and decision losses
Learning Reasoning Strategies in 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 |
|---|---|---|---|
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP · Nations | MRR: 0.658 -> 0.709 | Helped |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP · Nations | Hits@1: 0.493 -> 0.562 | Helped |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP · Nations | Hits@3: 0.781 -> 0.813 | Helped |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP · Nations | Hits@10: 0.985 -> 0.995 | Helped |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP Attention · Kinship | Hits@1: 0.642 -> 0.646 | Helped |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP Attention · UMLS | MRR: 0.857 -> 0.852 | Hurt |
| Use goal-conditioned, learned rule selection instead of GNTP rule selection | CTP versus GNTP Attention · Countries S2 | AUC-PR: 93.48 ± 3.29 -> 91.81 ± 1.07 | Hurt |
| Generate a small goal-conditioned rule set rather than iterate over all available rules | CTP · Nations | rules considered per goal: sixty rules -> two rules | Helped |
Closest papers by meaning
- Reviving DSP for Advanced Theorem Proving in the Era of Reasoning Models (NeurIPS 2025 · similarity 0.90)
- miniCTX: Neural Theorem Proving with (Long-)Contexts (ICLR 2025 · similarity 0.88)
- ProofAug: Efficient Neural Theorem Proving via Fine-grained Proof Structure Analysis (ICML 2025 · similarity 0.88)
- Learning Better Representations From Less Data For Propositional Satisfiability (NeurIPS 2024 · similarity 0.87)
- REFACTOR: Learning to Extract Theorems from Proofs (ICLR 2024 · similarity 0.87)
- Improving Soft Unification with Knowledge Graph Embedding Methods (ICML 2025 · similarity 0.87)
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.