Image de couverture de livre

LightTransfer: Your Long-Context LLM is Secretly a Hybrid Model with Effortless Adaptation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Zhang, Xuan, Zhang, Fengzhuo, Du, Cunxiao, Du, Chao, Pang, Tianyu, Gao, Wei, Lin, Min
Format:	Preprint
Publié:	2024
Sujets:	Computation and Language Artificial Intelligence Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification
par: Yang, Penghui, et autres
Publié: (2025)

When Attention Sink Emerges in Language Models: An Empirical View
par: Gu, Xiangming, et autres
Publié: (2024)

Demystifying the Slash Pattern in Attention: The Role of RoPE
par: Cheng, Yuan, et autres
Publié: (2026)

When Precision Meets Position: BFloat16 Breaks Down RoPE in Long-Context Training
par: Wang, Haonan, et autres
Publié: (2024)

Sparse-to-Dense: A Free Lunch for Lossless Acceleration of Video Understanding in LLMs
par: Zhang, Xuan, et autres
Publié: (2025)

Chain of Preference Optimization: Improving Chain-of-Thought Reasoning in LLMs
par: Zhang, Xuan, et autres
Publié: (2024)

Your Large Language Model is Secretly a Fairness Proponent and You Should Prompt it Like One
par: Li, Tianlin, et autres
Publié: (2024)

Improving Your Model Ranking on Chatbot Arena by Vote Rigging
par: Min, Rui, et autres
Publié: (2025)

Annealed Relaxation of Speculative Decoding for Faster Autoregressive Image Generation
par: Li, Xingyao, et autres
Publié: (2026)

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms
par: Hou, Yunlong, et autres
Publié: (2025)

SWIFT: On-the-Fly Self-Speculative Decoding for LLM Inference Acceleration
par: Xia, Heming, et autres
Publié: (2024)

Cheating Automatic LLM Benchmarks: Null Models Achieve High Win Rates
par: Zheng, Xiaosen, et autres
Publié: (2024)

A Closer Look at Machine Unlearning for Large Language Models
par: Yuan, Xiaojian, et autres
Publié: (2024)

Denial-of-Service Poisoning Attacks against Large Language Models
par: Gao, Kuofeng, et autres
Publié: (2024)

Muon Outperforms Adam in Tail-End Associative Memory Learning
par: Wang, Shuche, et autres
Publié: (2025)

Tutorial Proposal: Speculative Decoding for Efficient LLM Inference
par: Xia, Heming, et autres
Publié: (2025)

Flora: Effortless Context Construction to Arbitrary Length and Scale
par: Chen, Tianxiang, et autres
Publié: (2025)

Meta-Unlearning on Diffusion Models: Preventing Relearning Unlearned Concepts
par: Gao, Hongcheng, et autres
Publié: (2024)

Benchmarking Large Multimodal Models against Common Corruptions
par: Zhang, Jiawei, et autres
Publié: (2024)

Long-Context Language Modeling with Parallel Context Encoding
par: Yen, Howard, et autres
Publié: (2024)

Revisiting the Markov Property for Machine Translation
par: Du, Cunxiao, et autres
Publié: (2024)

Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses
par: Zheng, Xiaosen, et autres
Publié: (2024)

Scalable Token-Level Hallucination Detection in Large Language Models
par: Min, Rui, et autres
Publié: (2026)

Think in Parallel, Answer as One: Logit Averaging for Open-Ended Reasoning
par: Wang, Haonan, et autres
Publié: (2025)

LoraHub: Efficient Cross-Task Generalization via Dynamic LoRA Composition
par: Huang, Chengsong, et autres
Publié: (2023)

Latent Thinking Optimization: Your Latent Reasoning Language Model Secretly Encodes Reward Signals in Its Latent Thoughts
par: Du, Hanwen, et autres
Publié: (2025)

Test-Time Backdoor Attacks on Multimodal Large Language Models
par: Lu, Dong, et autres
Publié: (2024)

Your Language Model Can Secretly Write Like Humans: Contrastive Paraphrase Attacks on LLM-Generated Text Detectors
par: Fang, Hao, et autres
Publié: (2025)

HyLRA: Hybrid Layer Reuse Attention for Efficient Long-Context Inference
par: Ai, Xuan, et autres
Publié: (2026)

Purifying Large Language Models by Ensembling a Small Language Model
par: Li, Tianlin, et autres
Publié: (2024)

Language Models Can Learn from Verbal Feedback Without Scalar Rewards
par: Luo, Renjie, et autres
Publié: (2025)

LycheeDecode: Accelerating Long-Context LLM Inference via Hybrid-Head Sparse Decoding
par: Lin, Gang, et autres
Publié: (2026)

LongHeads: Multi-Head Attention is Secretly a Long Context Processor
par: Lu, Yi, et autres
Publié: (2024)

From Words to Actions: Unveiling the Theoretical Underpinnings of LLM-Driven Autonomous Systems
par: He, Jianliang, et autres
Publié: (2024)

Lifelong Safety Alignment for Language Models
par: Wang, Haoyu, et autres
Publié: (2025)

Your Mixture-of-Experts LLM Is Secretly an Embedding Model For Free
par: Li, Ziyue, et autres
Publié: (2024)

LongProc: Benchmarking Long-Context Language Models on Long Procedural Generation
par: Ye, Xi, et autres
Publié: (2025)

Efficiently Computing Susceptibility to Context in Language Models
par: Liu, Tianyu, et autres
Publié: (2024)

Bootstrapping Language Models with DPO Implicit Rewards
par: Chen, Changyu, et autres
Publié: (2024)

LLMSteer: Improving Long-Context LLM Inference by Steering Attention on Reused Contexts
par: Gu, Zhuohan, et autres
Publié: (2024)