Copertina

Why Attention Patterns Exist: A Unifying Temporal Perspective Analysis

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Yang, Qingyue, Wang, Jie, Li, Xing, Bai, Yinqi, Tong, Xialiang, Zhen, Huiling, Hao, Jianye, Yuan, Mingxuan, Li, Bin
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Computation and Language
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

AttentionPredictor: Temporal Patterns Matter for KV Cache Compression
di: Yang, Qingyue, et al.
Pubblicazione: (2025)

Accelerating Large Language Model Reasoning via Speculative Search
di: Wang, Zhihai, et al.
Pubblicazione: (2025)

ARS: Automatic Routing Solver with Large Language Models
di: Li, Kai, et al.
Pubblicazione: (2025)

Attention-Aware GNN-based Input Defense against Multi-Turn LLM Jailbreak
di: Huang, Zixuan, et al.
Pubblicazione: (2025)

Large Language Model for Multi-objective Evolutionary Optimization
di: Liu, Fei, et al.
Pubblicazione: (2023)

UDC: A Unified Neural Divide-and-Conquer Framework for Large-Scale Combinatorial Optimization Problems
di: Zheng, Zhi, et al.
Pubblicazione: (2024)

Preserving LLM Capabilities through Calibration Data Curation: From Analysis to Optimization
di: He, Bowei, et al.
Pubblicazione: (2025)

SVDq: 1.25-bit and 410x Key Cache Compression for LLM Attention
di: Yankun, Hong, et al.
Pubblicazione: (2025)

Beyond Speedup -- Utilizing KV Cache for Sampling and Reasoning
di: Xing, Zeyu, et al.
Pubblicazione: (2026)

Exploring Mathematical Extrapolation of Large Language Models with Synthetic Data
di: Li, Haolong, et al.
Pubblicazione: (2024)

Attention Needs to Focus: A Unified Perspective on Attention Allocation
di: Fu, Zichuan, et al.
Pubblicazione: (2026)

Why Is Spatial Reasoning Hard for VLMs? An Attention Mechanism Perspective on Focus Areas
di: Chen, Shiqi, et al.
Pubblicazione: (2025)

Why Softmax Attention Outperforms Linear Attention
di: Deng, Yichuan, et al.
Pubblicazione: (2023)

What Matters For Safety Alignment?
di: Li, Xing, et al.
Pubblicazione: (2026)

Attention Editing: A Versatile Framework for Cross-Architecture Attention Conversion
di: Cheng, Zhen, et al.
Pubblicazione: (2026)

A Systematic Survey on Large Language Models for Algorithm Design
di: Liu, Fei, et al.
Pubblicazione: (2024)

The Graph's Apprentice: Teaching an LLM Low Level Knowledge for Circuit Quality Estimation
di: Moravej, Reza, et al.
Pubblicazione: (2024)

EoH-S: Evolution of Heuristic Set using LLMs for Automated Heuristic Design
di: Liu, Fei, et al.
Pubblicazione: (2025)

Behavioral Fingerprinting of Large Language Models
di: Pei, Zehua, et al.
Pubblicazione: (2025)

Look Within, Why LLMs Hallucinate: A Causal Perspective
di: Li, He, et al.
Pubblicazione: (2024)

SwiftMem: Fast Agentic Memory via Query-aware Indexing
di: Tian, Anxin, et al.
Pubblicazione: (2026)

Mixed-R1: Unified Reward Perspective For Reasoning Capability in Multimodal Large Language Models
di: Xu, Shilin, et al.
Pubblicazione: (2025)

Existing LLMs Are Not Self-Consistent For Simple Tasks
di: Lin, Zhenru, et al.
Pubblicazione: (2025)

Unlocking the Secrets of Linear Complexity Sequence Model from A Unified Perspective
di: Qin, Zhen, et al.
Pubblicazione: (2024)

Slow Tuning and Low-Entropy Masking for Safe Chain-of-Thought Distillation
di: Ma, Ziyang, et al.
Pubblicazione: (2025)

Large Language Models Have Intrinsic Meta-Cognition, but Need a Good Lens
di: Ma, Ziyang, et al.
Pubblicazione: (2025)

KVTuner: Sensitivity-Aware Layer-Wise Mixed-Precision KV Cache Quantization for Efficient and Nearly Lossless LLM Inference
di: Li, Xing, et al.
Pubblicazione: (2025)

UniChange: Unifying Change Detection with Multimodal Large Language Model
di: Zhang, Xu, et al.
Pubblicazione: (2025)

Rethinking Creativity Evaluation: A Critical Analysis of Existing Creativity Evaluations
di: Lu, Li-Chun, et al.
Pubblicazione: (2025)

Does Understanding Inform Generation in Unified Multimodal Models? From Analysis to Path Forward
di: Niu, Yuwei, et al.
Pubblicazione: (2025)

Short Chains, Deep Thoughts: Balancing Reasoning Efficiency and Intra-Segment Capability via Split-Merge Optimization
di: Gui, Runquan, et al.
Pubblicazione: (2026)

AEIOU: A Unified Defense Framework against NSFW Prompts in Text-to-Image Models
di: Wang, Yiming, et al.
Pubblicazione: (2024)

TrimR: Verifier-based Training-Free Thinking Compression for Efficient Test-Time Scaling
di: Lin, Weizhe, et al.
Pubblicazione: (2025)

Scaling Linear Attention with Sparse State Expansion
di: Pan, Yuqi, et al.
Pubblicazione: (2025)

Scaling Up, Speeding Up: A Benchmark of Speculative Decoding for Efficient LLM Test-Time Scaling
di: Sun, Shengyin, et al.
Pubblicazione: (2025)

Attention2Probability: Attention-Driven Terminology Probability Estimation for Robust Speech-to-Text System
di: Du, Yanfan, et al.
Pubblicazione: (2025)

MVSS: A Unified Framework for Multi-View Structured Survey Generation
di: Liu, Yinqi, et al.
Pubblicazione: (2026)

HyperTree Planning: Enhancing LLM Reasoning via Hierarchical Thinking
di: Gui, Runquan, et al.
Pubblicazione: (2025)

Self-Improved Learning for Scalable Neural Combinatorial Optimization
di: Luo, Fu, et al.
Pubblicazione: (2024)

Fitness Landscape of Large Language Model-Assisted Automated Algorithm Search
di: Liu, Fei, et al.
Pubblicazione: (2025)