paper-with-me

Papers

Graph-to-Text Generation with Dynamic Structure Pruning

2022-09-15 · COLING 2022 10 · Liang Li, Ruiying Geng, Bowen Li, Can Ma, Yinliang Yue, Binhua Li, Yongbin Li

Most graph-to-text works are built on the encoder-decoder framework with cross-attention mechanism. Recent studies have shown that explicitly modeling the input graph structure can significantly improve the performance. However, the vanilla structural encoder cannot capture all specialized information in a single forward pass for all decoding steps, resulting in inaccurate semantic representations. Meanwhile, the input graph is flatted as an unordered sequence in the cross attention, ignoring the original graph structure. As a result, the obtained input graph context vector in the decoder may be flawed. To address these issues, we propose a Structure-Aware Cross-Attention (SACA) mechanism to re-encode the input graph representation conditioning on the newly generated context at each decoding step in a structure aware manner. We further adapt SACA and introduce its variant Dynamic Graph Pruning (DGP) mechanism to dynamically drop irrelevant nodes in the decoding process. We achieve new state-of-the-art results on two graph-to-text datasets, LDC2020T02 and ENT-DESC, with only minor increase on computational cost.

📄 PDF Abstract BibTeX arXiv:2209.07258

Code (0)

등록된 구현이 없습니다.

Tasks

DecoderText Generation

Methods 이 논문이 사용한 방법론

Pruning 설명 없음
AWARE We propose to theoretically and empirically examine the effect of incorporating weighting schemes into walk-aggregating GNNs. To this end, we propose a simple, interpretable, and…

Similar Papers 제목 키워드 기반

You Don't Need Pre-built Graphs for RAG: Retrieval Augmented Generation with Adaptive Reasoning Structures

2025-08-08 · Shengyuan Chen, Chuang Zhou, Zheng Yuan, Qinggang Zhang 외 arxiv

Large language models (LLMs) often suffer from hallucination, generating factually incorrect statements when handling questions beyond their knowledge and perception. Retrieval-augmented generation (RAG) addresses this b…

Modeling Unified Semantic Discourse Structure for High-quality Headline Generation

2024-03-23 · Minghui Xu, Hao Fei, Fei Li, Shengqiong Wu 외

Headline generation aims to summarize a long document with a short, catchy title that reflects the main idea. This requires accurately capturing the core document semantics, which is challenging due to the lengthy and ba…

Abstract Meaning RepresentationHeadline GenerationSentence

Pruning for Generalization: A Transfer-Oriented Spatiotemporal Graph Framework

2026-02-04 · Zihao Jing, Yuxi Long, Ganlin Feng arxiv

Multivariate time series forecasting in graph-structured domains is critical for real-world applications, yet existing spatiotemporal models often suffer from performance degradation under data scarcity and cross-domain …

Multivariate Time Series Forecasting

PathRAG: Pruning Graph-based Retrieval Augmented Generation with Relational Paths

2025-02-18 · BoYu Chen, Zirui Guo, Zidan Yang, Yuluo Chen 외

Retrieval-augmented generation (RAG) improves the response quality of large language models (LLMs) by retrieving knowledge from external databases. Typical RAG approaches split the text database into chunks, organizing t…

RAGRetrievalRetrieval-augmented Generation

Train Large, Deploy Compact: Structured Compression for Compact Low-Rank Adaptation

2025-09-30 · Xin Yu, Cong Xie, Ziyu Zhao, Tiantian Fan 외 arxiv

Low-rank adaptation (LoRA) has become a widely used paradigm for parameter-efficient fine-tuning of large language models, yet its representational capacity often lags behind full fine-tuning. Within the context of LoRA,…

parameter-efficient fine-tuningNatural Language UnderstandingMathematical ReasoningCode Generation