Research map / Large language models / Retrieval and knowledge
MAPS-KB: A Million-Scale Probabilistic Simile Knowledge Base
Qianyu He, Xintao Wang, Jiaqing Liang, Yanghua Xiao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 co-training across the like and be views instead of self-supervision on the like view | BERT_BASE simile detection · Wikipedia | F1: 81.49 -> 81.52 | Improved |
| Use co-training across the like and be views instead of self-supervision on the like view | BERT_BASE simile detection · UGC | F1: 80.66 -> 80.88 | Improved |
| Use constituency-parsing-based rules instead of regex rules | Component extraction · Hard topic extraction | F1: 52.19 -> 62.14 | Improved |
| Use constituency-parsing-based rules instead of regex rules | Component extraction · Hard vehicle extraction | F1: 65.73 -> 79.69 | Improved |
| Use constituency-parsing-based rules instead of regex rules | Component extraction · Easy vehicle extraction | F1: 90.54 -> 90.32 | No clear effect |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of the strongest listed baseline | Simile interpretation · Linguistics | MRR: 0.312 -> 0.392 | Improved |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of LMASKB | Simile interpretation · Quizzes | MRR: 0.342 -> 0.375 | Improved |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of LMASKB | Simile interpretation · General Corpus | MRR: 0.196 -> 0.200 | Improved |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of LMASKB | Simile generation · Linguistics | MRR: 0.095 -> 0.105 | Improved |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of LMASKB | Simile generation · Quizzes | MRR: 0.042 -> 0.132 | Improved |
| Use MAPS-KB probabilistic knowledge for candidate ranking instead of LMASKB | Simile generation · General Corpus | MRR: 0.022 -> 0.067 | Improved |
| Combine PLM generation with rule-based fallback rather than using the PLM method alone (ablation) | MAPS-KB writing polishing · Writing polishing | BLEU1: 11.99 -> 13.11 | Improved |
Closest papers by meaning
- MDK12-Bench: A Multi-Discipline Benchmark for Evaluating Reasoning in Multimodal Large Language Models (AAAI 2026 · similarity 0.85)
- MME-SCI: A Comprehensive and Challenging Science Benchmark for Multimodal Large Language Models (AAAI 2026 · similarity 0.84)
- Correlation Dimension of Autoregressive Large Language Models (NeurIPS 2025 · similarity 0.84)
- A Systematic Exploration of Knowledge Graph Alignment with Large Language Models in Retrieval Augmented Generation (AAAI 2025 · similarity 0.84)
- Journey to the Center of the Knowledge Neurons: Discoveries of Language-Independent Knowledge Neurons and Degenerate Knowledge Neurons (AAAI 2024 · similarity 0.84)
- Finding the Minimal Parameter Budget for Implicit Reasoning: A Data Complexity Driven Scaling Law for Language Models (ICML 2026 · similarity 0.84)
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.
