AI Paper Atlas

Research map / Large language models / Retrieval and knowledge

LLM-MatLogic: Executable Exchange Contracts for Knowledge-Graph Query Answering with Scoped Negation

Dezhuang Miao, Xiaoming Zhang, Bo Zhang, Yibin Du, Xiang Li, Ruilin Zeng, Yirui Qi

ICML 2026 · Large language models › Retrieval and knowledge · approach: Knowledge graph reasoning with LLMs

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 changedWhereMeasuredEffect
give oracle (known) negation constraints instead of specifier-generated (ablation)LLM-MatLogic · Five negation patterns (avg)MRR: 0.199 -> 0.369Improved
give oracle (known) negation constraints (ablation)LLM-MatLogic · Five negation patterns (avg)Hits@1: 0.181 -> 0.300Improved
give oracle constraints on 2in (ablation)LLM-MatLogic · 2inMRR: 0.162 -> 0.479Improved
give oracle constraints on pni (ablation)LLM-MatLogic · pniMRR: 0.197 -> 0.407Improved
use a stronger specifier LLM (DeepSeek-R1-1.5B to Llama-3-8B to GPT-3.5) (ablation)LLM-MatLogic · FB15k-237 3inMRR: 1.8 -> 4.9 -> 12.7Improved
use a stronger specifier LLM (ablation)LLM-MatLogic · FB15k-237 2inMRR: 2.4 -> 5.5 -> 11.4Improved
use LLM-MatLogic instead of MFC on composition queriesLLM-MatLogic · FB15k-237 2pMRR: 35.2 -> 58.2Improved
use LLM-MatLogic instead of MFC on NELL995 2pLLM-MatLogic · NELL995 2pMRR: 39.6 -> 64.5Improved
use LLM-MatLogic instead of LARK on 2in/3in/pniLLM-MatLogic · FB15k-237 negationMRR: 7.5/4.8/3.6 -> 11.4/12.7/10.9Improved
use LLM-MatLogic instead of LARK on inp/pinLLM-MatLogic · FB15k-237 negationMRR: 23.3 -> 15.1 (inp)Hurt
use LLM-MatLogic instead of MFC on CWQLLM-MatLogic · CWQHits@1: 62.8 -> 63.5Improved
use LLM-MatLogic instead of MFC on WebQSPLLM-MatLogic · WebQSPHits@1: 78.9 -> 76.7Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy