Project SHADOW: Symbolic Higher-order Associative Deductive reasoning On Wikidata using LM probing
arXiv:2408.14849v2h-index: 5
AI Analysis
This work addresses knowledge base completion for AI systems, but it is incremental as it builds on existing language model methods.
The paper tackled the problem of knowledge base construction by introducing SHADOW, a fine-tuned language model for associative deductive reasoning, and achieved a 20% improvement over the baseline with an F1 score of 68.72% on the LM-KBC 2024 challenge.
We introduce SHADOW, a fine-tuned language model trained on an intermediate task using associative deductive reasoning, and measure its performance on a knowledge base construction task using Wikidata triple completion. We evaluate SHADOW on the LM-KBC 2024 challenge and show that it outperforms the baseline solution by 20% with a F1 score of 68.72%.