Local-to-Global Sentence-Level Graph Reranking for Scientific Synthesis
2026-08-31 • Information Retrieval
Information Retrieval
AI summaryⓘ
The authors focus on improving how computers gather and organize information from multiple scientific papers to answer complex research questions. They point out that current methods rank information in large chunks, which can hide important details, and often look at each piece of information separately without considering how they relate. To fix this, the authors created LoG-Reranker, a system that first looks closely at individual sentences to find the most relevant ones, then connects these sentences in a graph to see how they relate overall. This approach leads to better organization and more reliable information for generating thorough scientific answers.
retrieval-augmented synthesisscientific rerankingpassage-level rerankersentence-level graphlocal-to-global scoringquery-candidate scoringinformation groundingcomprehensive synthesisnatural language generationsentence graph
Authors
Zheng Dou, Zhao Zhang, Hao Geng, Ningjing Wang, Deqing Wang
Abstract
Retrieval-augmented scientific synthesis aims to answer complex research questions by integrating information from multiple papers into comprehensive and well-grounded responses. Since the generator can only synthesize the information selected and organized by the reranker, the quality of the generated synthesis depends critically on the reranked results. However, most rerankers operate at the passage level, which leaves key methodological, empirical, and comparative information buried in long and flat contexts, weakening the grounding of generated claims. Moreover, existing rerankers mainly rely on independent query-candidate scoring which overlooks complementary, contextual, and contrasting relations across scientific candidates, limiting information coverage and the comprehensiveness of the resulting synthesis. To address these limitations, we propose LoG-Reranker, a local-to-global sentence-level graph reranking framework for scientific synthesis. LoG-Reranker performs role-aware local scoring to identify fine-grained, query-relevant sentences and then models their relations on a sentence graph across the candidate set to globally refine sentence rankings. Top-ranked sentences and their connected neighbors are organized into a structured input context for generator to produce more grounded and comprehensive synthesis.Extensive experiments on scientific synthesis and reranking benchmarks show that LoG-Reranker consistently outperforms competitive rerankers, yielding more reliable rankings and improving the quality of generated synthesis.