paper-with-me

Papers

Long Context Scaling: Divide and Conquer via Multi-Agent Question-driven Collaboration

2025-05-27 · Sibo Xiao, Zixin Lin, Wenyang Gao, Yue Zhang

Processing long contexts has become a critical capability for modern large language models (LLMs). Existing works leverage agent-based divide-and-conquer methods for processing long contexts. But these methods face crucial limitations, including prohibitive accumulated latency and amplified information loss from excessive agent invocations, and the disruption of inherent textual dependencies by immoderate partitioning. In this paper, we propose a novel multi-agent framework XpandA (Expand-Agent) coupled with question-driven workflow and dynamic partitioning for robust long-context processing. XpandA overcomes these limitations through: 1) dynamic partitioning of long texts, which adaptively modulates the filling rate of context windows for input sequences of vastly varying lengths; 2) question-guided protocol to update flat information ensembles within centralized shared memory, constructing consistent inter-agent knowledge across partitions; and 3) selectively replaying specific partitions based on the state-tracking of question-information couples to promote the resolution of inverted-order structures across partitions (e.g., flashbacks). We perform a comprehensive evaluation of XpandA on multiple long-context benchmarks with length varying from 1k to 1M, demonstrating XpandA's feasibility for processing ultra-long sequences and its significant effectiveness in enhancing the long-context capabilities of various LLMs by achieving 20\% improvements and 1.5x inference speedup over baselines of full-context, RAG and previous agent-based methods.

📄 PDF Abstract BibTeX arXiv:2505.20625

Code (0)

등록된 구현이 없습니다.

Tasks

RAG

Methods 이 논문이 사용한 방법론

BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Attention 설명 없음
WordPiece 설명 없음
Multi-Head Attention 설명 없음
BART BART is a denoising autoencoder for pretraining sequence-to-sequence models. It is trained by (1) corrupting text…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…

Similar Papers 제목 키워드 기반

DCIS: Efficient Length Extrapolation of LLMs via Divide-and-Conquer Scaling Factor Search

2024-12-25 · Lei Yang, Shaoyang Xu, Deyi Xiong

Large language models (LLMs) based on the Transformer architecture usually have their context length limited due to the high training cost. Recent advancements extend the context window by adjusting the scaling factors o…

DCARL: A Divide-and-Conquer Framework for Autoregressive Long-Trajectory Video Generation

2026-03-25 · Junyi Ouyang, Wenbin Teng, Gonglin Chen, Yajie Zhao 외 arxiv

Long-trajectory video generation is a crucial yet challenging task for world modeling primarily due to the limited scalability of existing video diffusion models (VDMs). Autoregressive models, while offering infinite rol…

Video Generation

Divide-and-conquer methods for big data analysis

2021-02-22 · Xueying Chen, Jerry Q. Cheng, Min-ge Xie

In the context of big data analysis, the divide-and-conquer methodology refers to a multiple-step process: first splitting a data set into several smaller ones; then analyzing each set separately; finally combining resul…

Two provably consistent divide and conquer clustering algorithms for large networks

2017-08-18 · Soumendu Sundar Mukherjee, Purnamrita Sarkar, Peter J. Bickel

In this article, we advance divide-and-conquer strategies for solving the community detection problem in networks. We propose two algorithms which perform clustering on a number of small subgraphs and finally patches the…

ClusteringCommunity DetectionVocal Bursts Valence Prediction

Recursive Agent Optimization

2026-05-07 · Apurva Gandhi, Satyaki Chakraborty, Xiangjun Wang, Aviral Kumar 외 arxiv

We introduce Recursive Agent Optimization (RAO), a reinforcement learning approach for training recursive agents: agents that can spawn and delegate sub-tasks to new instantiations of themselves recursively. Recursive ag…

Reinforcement Learning