Imagen de Portada

Length Value Model: Scalable Value Pretraining for Token-Level Length Modeling

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Zhang, Zhen, Yang, Changyi, Xia, Zijie, Yang, Zhen, Liu, Chengzhi, Weng, Zhaotiao, Liu, Yepeng, Chen, Haobo, Pan, Jin, Zhao, Chenyang, Bu, Yuheng, Patel, Alkesh, Gan, Zhe, Wang, Xin Eric
Formato:	Preprint
Publicado:	2026
Materias:	Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

Adaptive Text Watermark for Large Language Models
por: Liu, Yepeng, et al.
Publicado: (2024)

In-Context Watermarks for Large Language Models
por: Liu, Yepeng, et al.
Publicado: (2025)

Advancing Egocentric Video Question Answering with Multimodal Large Language Models
por: Patel, Alkesh, et al.
Publicado: (2025)

A Reinforcement Learning Framework for Robust and Secure LLM Watermarking
por: An, Li, et al.
Publicado: (2025)

Defending LLM Watermarking Against Spoofing Attacks with Contrastive Representation Learning
por: An, Li, et al.
Publicado: (2025)

Dataset Protection via Watermarked Canaries in Retrieval-Augmented LLMs
por: Liu, Yepeng, et al.
Publicado: (2025)

Length-MAX Tokenizer for Language Models
por: Dong, Dong, et al.
Publicado: (2025)

A Benchmark for Databases with Varying Value Lengths
por: Liyanage, Danushka, et al.
Publicado: (2025)

Iterative Length-Regularized Direct Preference Optimization: A Case Study on Improving 7B Language Models to GPT-4 Level
por: Liu, Jie, et al.
Publicado: (2024)

Reflection Pretraining Enables Token-Level Self-Correction in Biological Sequence Models
por: Zhang, Xiang, et al.
Publicado: (2025)

Efficient Pretraining Length Scaling
por: Wu, Bohong, et al.
Publicado: (2025)

Joint Optimization of Pilot Length, Pilot Assignment, and Power Allocation for Cell-free MIMO Systems with Graph Neural Networks
por: Peng, Yao, et al.
Publicado: (2025)

Distributional Information Embedding: A Framework for Multi-bit Watermarking
por: He, Haiyun, et al.
Publicado: (2025)

Theoretically Grounded Framework for LLM Watermarking: A Distribution-Adaptive Approach
por: He, Haiyun, et al.
Publicado: (2024)

An Algorithm for Computing the Capacity of Symmetrized KL Information for Discrete Channels
por: Chen, Haobo, et al.
Publicado: (2024)

Reward Models Inherit Value Biases from Pretraining
por: Christian, Brian, et al.
Publicado: (2026)

Length Desensitization in Direct Preference Optimization
por: Liu, Wei, et al.
Publicado: (2024)

Can GRPO Help LLMs Transcend Their Pretraining Origin?
por: Ni, Kangqi, et al.
Publicado: (2025)

Position: LLM Watermarking Should Align Stakeholders' Incentives for Practical Adoption
por: Liu, Yepeng, et al.
Publicado: (2025)

ConvexBench: Can LLMs Recognize Convex Functions?
por: Liu, Yepeng, et al.
Publicado: (2026)

Length-Factoriality and Pure Irreducibility
por: Bu, Alan, et al.
Publicado: (2022)

TokenPure: Watermark Removal through Tokenized Appearance and Structural Guidance
por: Yang, Pei, et al.
Publicado: (2025)

Auditing Agent Harness Safety
por: Liu, Chengzhi, et al.
Publicado: (2026)

Pretrain Value, Not Reward: Decoupled Value Policy Optimization
por: Huang, Chenghua, et al.
Publicado: (2025)

Megalodon: Efficient LLM Pretraining and Inference with Unlimited Context Length
por: Ma, Xuezhe, et al.
Publicado: (2024)

Various Lengths, Constant Speed: Efficient Language Modeling with Lightning Attention
por: Qin, Zhen, et al.
Publicado: (2024)

Improving Variable-Length Generation in Diffusion Language Models via Length Regularization
por: Cheng, Zicong, et al.
Publicado: (2026)

Do Language Models Think Consistently? A Study of Value Preferences Across Varying Response Lengths
por: Nair, Inderjeet, et al.
Publicado: (2025)

On the Value of Tokeniser Pretraining in Physics Foundation Models
por: Sotoudeh, Hadi, et al.
Publicado: (2026)

LLM Serving Optimization with Variable Prefill and Decode Lengths
por: Wang, Meixuan, et al.
Publicado: (2025)

Fundamental Trade-Offs in Multi-Bit Watermarking of Stochastic Processes
por: He, Haiyun, et al.
Publicado: (2026)

TokenShapley: Token Level Context Attribution with Shapley Value
por: Xiao, Yingtai, et al.
Publicado: (2025)

Finite Correlation Length Scaling of Disorder Parameter at Quantum Criticality
por: Xu, Wen-Tao, et al.
Publicado: (2024)

Sets of Lengths of Integer-Valued Polynomials on Prime Ideals of Principal Ideal Domains
por: Kansiime, Zaituni, et al.
Publicado: (2026)

An Extreme Value Theory Approach for Understanding Queue Length Dynamics in Adaptive Corridors
por: Mustavee, Shakib, et al.
Publicado: (2024)

Length Matters: Length-Aware Transformer for Temporal Sentence Grounding
por: Wang, Yifan, et al.
Publicado: (2025)

Length-Unbiased Sequence Policy Optimization: Revealing and Controlling Response Length Variation in RLVR
por: Liu, Fanfan, et al.
Publicado: (2026)

DVMap: Fine-Grained Pluralistic Value Alignment via High-Consensus Demographic-Value Mapping
por: Zhu, Pengyun, et al.
Publicado: (2026)

Learning Variable-Length Tokenization for Generative Recommendation
por: Wang, Minhao, et al.
Publicado: (2026)

Tokenizing Semantic Segmentation with Run Length Encoding
por: Singh, Abhineet, et al.
Publicado: (2026)