logical-reasoning

KAG is a logical form-guided reasoning and retrieval framework based on OpenSPG engine and LLMs. It is used to build logical reasoning and factual Q&A solutions for professional domain knowledge bases. It can effectively overcome the shortcomings of the traditional RAG vector similarity calculation model.

The Art of Problem-Solving in Software Engineering: How to Make MySQL Better

[ACL 2023] Reasoning with Language Model Prompting: A Survey

Recent Papers including Neural Symbolic Reasoning, Logical Reasoning, Visual Reasoning, planning and any other topics connecting deep learning and reasoning

Inductive relation prediction by subgraph reasoning, ICML'20

The official repo of SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond

On Memorization of Large Language Models in Logical Reasoning

Logic Circuits from the Juice library

Official Repository: A Comprehensive Benchmark for Logical Reasoning in MLLMs

[ICML 2023] Answering Complex Logical Queries on Knowledge Graphs via Query Computation Tree Optimization

[SIGIR 2022] The implementation of Logiformer

Evaluation on Logical Reasoning and Abstract Reasoning Challenges

[AAAI 2023] Official resources of "NQE: N-ary Query Embedding for Complex Query Answering over Hyper-relational Knowledge Graphs".

[EMNLP 2024] A Peek into Token Bias: Large Language Models Are Not Yet Genuine Reasoners

Repo for paper "IDOL: Indicator-oriented Logic Pre-training for Logical Reasoning" accepted to the Findings of ACL 2023

Official code for "Divide and Translate: Compositional First-Order Logic Translation and Verification for Complex Logical Reasoning", ICLR 2025.

The source code for Abstract Meaning Representation-Based Logic-Driven Data Augmentation for Logical Reasoning. #1 on the ReClor Leaderboard and we are the first group scored above 90% on the hidden test set around the world. The paper has been accepted by the Findings of ACL-24.

[NeurIPS2023] LoRA: A Logical Reasoning Augmented Dataset for Visual Question Answering

Understanding Expressivity of GNN in Rule Learning. ICLR 2024

An evaluation dataset comprising of 274 grid-based puzzles with different complexities