# HSGM: Hierarchical Segment-Graph Memory for Scalable Long-Text Semantics

arXiv: https://arxiv.org/abs/2509.18168 · HTML: https://mlsanity.com/paper/2509.18168/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 17 of its 29 references that are in the map.
- First posted on arXiv 17 Sep 2025.

## Builds on (most cited first, 17 of 17)

- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Retrieval-Augmented Generation for Knowledge-Intensive NLP Tasks](https://mlsanity.com/paper/2005.11401/) (2020)
- [Inductive Representation Learning on Large Graphs](https://mlsanity.com/paper/1706.02216/) (2017)
- [Dense Passage Retrieval for Open-Domain Question Answering](https://mlsanity.com/paper/2004.04906/) (2020)
- [Longformer: The Long-Document Transformer](https://mlsanity.com/paper/2004.05150/) (2020)
- [Big Bird: Transformers for Longer Sequences](https://mlsanity.com/paper/2007.14062/) (2020)
- [Reformer: The Efficient Transformer](https://mlsanity.com/paper/2001.04451/) (2020)
- [LongBench: A Bilingual, Multitask Benchmark for Long Context Understanding](https://mlsanity.com/paper/2308.14508/) (2023)
- [REPLUG: Retrieval-Augmented Black-Box Language Models](https://mlsanity.com/paper/2301.12652/) (2023)
- [The NarrativeQA Reading Comprehension Challenge](https://mlsanity.com/paper/1712.07040/) (2017)
- [LongT5: Efficient Text-To-Text Transformer for Long Sequences](https://mlsanity.com/paper/2112.07916/) (2021)
- [SGPT: GPT Sentence Embeddings for Semantic Search](https://mlsanity.com/paper/2202.08904/) (2022)
- [Constructing Datasets for Multi-hop Reading Comprehension Across Documents](https://mlsanity.com/paper/1710.06481/) (2017)
- [Adaptive Attention Span in Transformers](https://mlsanity.com/paper/1905.07799/) (2019)
- [DynGEM: Deep Embedding Method for Dynamic Graphs](https://mlsanity.com/paper/1805.11273/) (2018)
- [An Exploration of Hierarchical Attention Transformers for Efficient Long Document Classification](https://mlsanity.com/paper/2210.05529/) (2022)
- [Extreme Multi-Label Legal Text Classification: A case study in EU Legislation](https://mlsanity.com/paper/1905.10892/) (2019)

## Built on by (most cited first, 5 of 5)

- [AdaCorrection: Adaptive Offset Cache Correction for Accurate Diffusion Transformers](https://mlsanity.com/paper/2602.13357/) (2026)
- [Accelerating Frequency Domain Diffusion Models with Error-Feedback Event-Driven Caching](https://mlsanity.com/paper/2604.22901/) (2026)
- [Thoughts-as-Planning: Latent World Models for Chain-of-Thoughts Optimization via Reinforcement Planning](https://mlsanity.com/paper/2605.28842/) (2026)
- [RUBRIC: Realism--Utility Balanced Ranking for Imbalanced Classification](https://mlsanity.com/paper/2607.09816/) (2026)
- [Cognitive Load Traces as Symbolic and Visual Accounts of Deep Model Cognition](https://mlsanity.com/paper/2510.15980/) (2025)

## Abstract (arXiv)

Semantic parsing of long documents remains challenging due to quadratic growth in pairwise composition and memory requirements. We introduce Hierarchical Segment-Graph Memory (HSGM), a novel framework that decomposes an input of length N into M meaningful segments, constructs Local Semantic Graphs on each segment, and extracts compact summary nodes to form a Global Graph Memory. HSGM supports incremental updates -- only newly arrived segments incur local graph construction and summary-node integration -- while Hierarchical Query Processing locates relevant segments via top-K retrieval over summary nodes and then performs fine-grained reasoning within their local graphs. Theoretically, HSGM reduces worst-case complexity from O(N2) to O!left(N,k + (N/k)2right), with segment size k ll N, and we derive Frobenius-norm bounds on the approximation error introduced by node summarization and sparsification thresholds. Empirically, on three benchmarks -- long-document AMR parsing, segment-level semantic role labeling (OntoNotes), and legal event extraction -- HSGM achieves 2--4× inference speedup, >60% reduction in peak memory, and ≥ 95% of baseline accuracy. Our approach unlocks scalable, accurate semantic modeling for ultra-long texts, enabling real-time and resource-constrained NLP applications.