Imagen de Portada

SpecBound: Adaptive Bounded Self-Speculation with Layer-wise Confidence Calibration

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Wen, Zhuofan, Feng, Yang
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language Artificial Intelligence Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Speculative Decoding with CTC-based Draft Model for LLM Inference Acceleration
por: Wen, Zhuofan, et al.
Publicado: (2024)

SpecDec++: Boosting Speculative Decoding via Adaptive Candidate Lengths
por: Huang, Kaixuan, et al.
Publicado: (2024)

HiSpec: Hierarchical Speculative Decoding for LLMs
por: Kumar, Avinash, et al.
Publicado: (2025)

SpecExit: Accelerating Large Reasoning Model via Speculative Exit
por: Yang, Rubing, et al.
Publicado: (2025)

DistillSpec: Improving Speculative Decoding via Knowledge Distillation
por: Zhou, Yongchao, et al.
Publicado: (2023)

LongSpec: Long-Context Lossless Speculative Decoding with Efficient Drafting and Verification
por: Yang, Penghui, et al.
Publicado: (2025)

ML-SpecQD: Multi-Level Speculative Decoding with Quantized Drafts
por: Georganas, Evangelos, et al.
Publicado: (2025)

SpecForge: A Flexible and Efficient Open-Source Training Framework for Speculative Decoding
por: Li, Shenggui, et al.
Publicado: (2026)

Token-Driven GammaTune: Adaptive Calibration for Enhanced Speculative Decoding
por: Gautam, Aayush, et al.
Publicado: (2025)

DynaSpec: Context-aware Dynamic Speculative Sampling for Large-Vocabulary Language Models
por: Zhang, Jinbin, et al.
Publicado: (2025)

Confidence-Modulated Speculative Decoding for Large Language Models
por: Sen, Jaydip, et al.
Publicado: (2025)

FR-Spec: Accelerating Large-Vocabulary Language Models via Frequency-Ranked Speculative Sampling
por: Zhao, Weilin, et al.
Publicado: (2025)

LayerSkip: Enabling Early Exit Inference and Self-Speculative Decoding
por: Elhoushi, Mostafa, et al.
Publicado: (2024)

SpecKV: Adaptive Speculative Decoding with Compression-Aware Gamma Selection
por: Shukla, Shikhar
Publicado: (2026)

KnapSpec: Self-Speculative Decoding via Adaptive Layer Selection as a Knapsack Problem
por: Cha, Seongjin, et al.
Publicado: (2026)

Decoupling Reasoning and Confidence: Resurrecting Calibration in Reinforcement Learning from Verifiable Rewards
por: Ma, Zhengzhao, et al.
Publicado: (2026)

ConfLayers: Adaptive Confidence-based Layer Skipping for Self-Speculative Decoding
por: Amer, Walaa, et al.
Publicado: (2026)

Adaptive-Boundary-Clipping GRPO: Ensuring Bounded Ratios for Stable and Generalizable Training
por: Liu, Chi, et al.
Publicado: (2026)

Calibration Across Layers: Understanding Calibration Evolution in LLMs
por: Joshi, Abhinav, et al.
Publicado: (2025)

Confidence-aware Self-Semantic Distillation on Knowledge Graph Embedding
por: Liu, Yichen, et al.
Publicado: (2022)

MICE for CATs: Model-Internal Confidence Estimation for Calibrating Agents with Tools
por: Subramani, Nishant, et al.
Publicado: (2025)

Self-Speculative Biased Decoding for Faster Re-Translation
por: Zeng, Linxiao, et al.
Publicado: (2025)

LEAP: Layer-wise Exit-Aware Pretraining for Efficient Transformer Inference
por: Kapadia, Shashank, et al.
Publicado: (2026)

MaskPrune: Mask-based LLM Pruning for Layer-wise Uniform Structures
por: Qin, Jiayu, et al.
Publicado: (2025)

SpecExtend: A Drop-in Enhancement for Speculative Decoding of Long Sequences
por: Cha, Jungyoub, et al.
Publicado: (2025)

Entropy-Adaptive Fine-Tuning: Resolving Confident Conflicts to Mitigate Forgetting
por: Diao, Muxi, et al.
Publicado: (2026)

Self-Guided Process Reward Optimization with Redefined Step-wise Advantage for Process Reinforcement Learning
por: Fei, Wu, et al.
Publicado: (2025)

CopySpec: Accelerating LLMs with Speculative Copy-and-Paste Without Compromising Quality
por: Dumitru, Razvan-Gabriel, et al.
Publicado: (2025)

ConfSpec: Efficient Step-Level Speculative Reasoning via Confidence-Gated Verification
por: Liu, Siran, et al.
Publicado: (2026)

SLaNC: Static LayerNorm Calibration
por: Salmani, Mahsa, et al.
Publicado: (2024)

BanditSpec: Adaptive Speculative Decoding via Bandit Algorithms
por: Hou, Yunlong, et al.
Publicado: (2025)

SaySelf: Teaching LLMs to Express Confidence with Self-Reflective Rationales
por: Xu, Tianyang, et al.
Publicado: (2024)

Certified Robustness Under Bounded Levenshtein Distance
por: Rocamora, Elias Abad, et al.
Publicado: (2025)

AdaSPEC: Selective Knowledge Distillation for Efficient Speculative Decoders
por: Hu, Yuezhou, et al.
Publicado: (2025)

Graph-based Confidence Calibration for Large Language Models
por: Li, Yukun, et al.
Publicado: (2024)

Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs
por: Bhattacharyya, Sree, et al.
Publicado: (2026)

AdaptiveStep: Automatically Dividing Reasoning Step through Model Confidence
por: Liu, Yuliang, et al.
Publicado: (2025)

Calibrating Language Models with Adaptive Temperature Scaling
por: Xie, Johnathan, et al.
Publicado: (2024)

Online Speculative Decoding
por: Liu, Xiaoxuan, et al.
Publicado: (2023)

CAS-Spec: Cascade Adaptive Self-Speculative Decoding for On-the-Fly Lossless Inference Acceleration of LLMs
por: Ning, Zhiyuan, et al.
Publicado: (2025)