Cover Image

From Similarity to Structure: Training-free LLM Context Compression with Hybrid Graph Priors

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Zhou, Yitian, Zhang, Chaoning, Zhang, Jiaquan, Huang, Zhenzhen, Guo, Jinyu, Bae, Sung-Ho, Lee, Lik-Hang, Qin, Caiyan, Yang, Yang
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

DASH-KV: Accelerating Long-Context LLM Inference via Asymmetric KV Cache Hashing
by: Guo, Jinyu, et al.
Published: (2026)

Small Language Model Helps Resolve Semantic Ambiguity of LLM Prompt
by: Huang, Zhenzhen, et al.
Published: (2026)

TDA-RC: Task-Driven Alignment for Knowledge-Based Reasoning Chains in Large Language Models
by: Zhang, Jiaquan, et al.
Published: (2026)

Rethinking Input Domains in Physics-Informed Neural Networks via Geometric Compactification Mappings
by: Huang, Zhenzhen, et al.
Published: (2026)

Experience Transfer for Multimodal LLM Agents in Minecraft Game
by: Li, Chenghao, et al.
Published: (2026)

Lightweight LLM Agent Memory with Small Language Models
by: Zhang, Jiaquan, et al.
Published: (2026)

Optimizing Soft Prompt Tuning via Structural Evolution
by: Huang, Zhenzhen, et al.
Published: (2026)

LLaVA-FA: Learning Fourier Approximation for Compressing Large Multimodal Models
by: Zheng, Pengcheng, et al.
Published: (2026)

Efficient and Interpretable Multi-Agent LLM Routing via Ant Colony Optimization
by: Wang, Xudong, et al.
Published: (2026)

Generative AI meets 3D: A Survey on Text-to-3D in AIGC Era
by: Li, Chenghao, et al.
Published: (2023)

Learning Global Hypothesis Space for Enhancing Synergistic Reasoning Chain
by: Zhang, Jiaquan, et al.
Published: (2026)

CAP-CoT: Cycle Adversarial Prompt for Improving Chain of Thoughts in LLM Reasoning
by: Chen, Shuxu, et al.
Published: (2026)

Fast SAM2 with Text-Driven Token Pruning
by: Mandal, Avilasha, et al.
Published: (2025)

Geometric Neural Operators via Lie Group-Constrained Latent Dynamics
by: Zhang, Jiaquan, et al.
Published: (2026)

GraphMind: Theorem Selection and Conclusion Generation Framework with Dynamic GNN for LLM Reasoning
by: Li, Yutong, et al.
Published: (2025)

Autoregression-Free Neural Operators for Time-Dependent PDEs
by: Zhang, Jiaquan, et al.
Published: (2026)

Compression in 3D Gaussian Splatting: A Survey of Methods, Trends, and Future Directions
by: Ali, Muhammad Salman, et al.
Published: (2025)

A Survey on Segment Anything Model (SAM): Vision Foundation Model Meets Prompt Engineering
by: Zhang, Chaoning, et al.
Published: (2023)

Language-Guided Token Compression with Reinforcement Learning in Large Vision-Language Models
by: Cao, Sihan, et al.
Published: (2026)

Topology-Aware Layer Pruning for Large Vision-Language Models
by: Zheng, Pengcheng, et al.
Published: (2026)

Transforming External Knowledge into Triplets for Enhanced Retrieval in RAG of LLMs
by: Wang, Xudong, et al.
Published: (2026)

Practical No-box Adversarial Attacks with Training-free Hybrid Image Transformation
by: Zhang, Qilong, et al.
Published: (2022)

Text summarization via global structure awareness
by: Zhang, Jiaquan, et al.
Published: (2026)

Weak-Link Optimization for Multi-Agent Reasoning and Collaboration
by: Bian, Haoyu, et al.
Published: (2026)

vGamba: Attentive State Space Bottleneck for efficient Long-range Dependencies in Visual Recognition
by: Haruna, Yunusa, et al.
Published: (2025)

Understanding Chain-of-Thought in Large Language Models via Topological Data Analysis
by: Li, Chenghao, et al.
Published: (2025)

Syzygy of Thoughts: Improving LLM CoT with the Minimal Free Resolution
by: Li, Chenghao, et al.
Published: (2025)

Towards Understanding Dual BN In Hybrid Adversarial Training
by: Zhang, Chenshuang, et al.
Published: (2024)

Unconstrained Multi-view Human Pose Estimation with Algebraic Priors
by: Qin, Xiaolin, et al.
Published: (2026)

The Shadow Boss: Identifying Atomized Manipulations in Agentic Employment of XR Users using Scenario Constructions
by: Lee, Lik-Hang
Published: (2026)

Immunizing 3D Gaussian Generative Models Against Unauthorized Fine-Tuning via Attribute-Space Traps
by: Zhang, Jianwei, et al.
Published: (2026)

RCP: Representation Consistency Pruner for Mitigating Distribution Shift in Large Vision-Language Models
by: Zhang, Jianwei, et al.
Published: (2026)

Relaxing Anchor-Frame Dominance for Mitigating Hallucinations in Video Large Language Models
by: Liu, Zijian, et al.
Published: (2026)

StructComp: Substituting Propagation with Structural Compression in Training Graph Contrastive Learning
by: Zhang, Shengzhong, et al.
Published: (2023)

Post Training Quantization for Efficient Dataset Condensation
by: Tran, Linh-Tam, et al.
Published: (2026)

Efficient RAG with Intent-Aware Retrieval and Semantics-Preserving Chunking
by: Puspitasari, Fachrina Dewi, et al.
Published: (2026)

Sora as a World Model? A Complete Survey on Text-to-Video Generation
by: Puspitasari, Fachrina Dewi, et al.
Published: (2024)

Hybrid Attribution Priors for Explainable and Robust Model Training
by: Zhang, Zhuoran, et al.
Published: (2025)

Compressing KV Cache for Long-Context LLM Inference with Inter-Layer Attention Similarity
by: Ma, Da, et al.
Published: (2024)

InPK: Infusing Prior Knowledge into Prompt for Vision-Language Models
by: Zhou, Shuchang, et al.
Published: (2025)