Cover Image

SpanNorm: Reconciling Training Stability and Performance in Deep Transformers

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Wang, Chao, Li, Bei, Zhang, Jiaqi, Liu, Xinyu, Fan, Yuchun, Lyu, Linkun, Chen, Xin, Wang, Jingang, Xiao, Tong, Pei, Peng, Cai, Xunliang
Format:	Preprint
Published:	2026
Subjects:	Computation and Language Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

LANG: Reinforcement Learning for Multilingual Reasoning with Language-Adaptive Hint Guidance
by: Fan, Yuchun, et al.
Published: (2026)

Predictor-Corrector Enhanced Transformers with Exponential Moving Average Coefficient Learning
by: Li, Bei, et al.
Published: (2024)

Speculative Decoding via Early-exiting for Faster LLM Inference with Thompson Sampling Control Mechanism
by: Liu, Jiahao, et al.
Published: (2024)

Ltri-LLM: Streaming Long Context Inference for LLMs with Training-Free Dynamic Triangular Attention Pattern
by: Tang, Hongyin, et al.
Published: (2024)

GradPower: Powering Gradients for Faster Language Model Pre-Training
by: Wang, Jinbo, et al.
Published: (2025)

MTR-Suite: A Framework for Evaluating and Synthesizing Conversational Retrieval Benchmarks
by: Ruan, Junhao, et al.
Published: (2026)

Causal Autoregressive Diffusion Language Model
by: Ruan, Junhao, et al.
Published: (2026)

Earlier Tokens Contribute More: Learning Direct Preference Optimization From Temporal Decay Perspective
by: Shao, Ruichen, et al.
Published: (2025)

Beyond the Known: Investigating LLMs Performance on Out-of-Domain Intent Detection
by: Wang, Pei, et al.
Published: (2024)

Scaling Embeddings Outperforms Scaling Experts in Language Models
by: Liu, Hong, et al.
Published: (2026)

ReMamba: Equip Mamba with Effective Long-Sequence Modeling
by: Yuan, Danlong, et al.
Published: (2024)

FIRE: Flexible Integration of Data Quality Ratings for Effective Pre-Training
by: Xu, Liangyu, et al.
Published: (2025)

Large-Scale Diverse Synthesis for Mid-Training
by: Zhang, Xuemiao, et al.
Published: (2025)

OneCAT: Decoder-Only Auto-Regressive Model for Unified Understanding and Generation
by: Li, Han, et al.
Published: (2025)

Sub-Scaling Laws: On the Role of Data Density and Training Strategies in LLMs
by: Chen, Zhengyu, et al.
Published: (2025)

Scaling and Transferability of Annealing Strategies in Large Language Model Training
by: Wang, Siqi, et al.
Published: (2025)

Libra: Assessing and Improving Reward Model by Learning to Think
by: Zhou, Meng, et al.
Published: (2025)

Dynamic Fisher-weighted Model Merging via Bayesian Optimization
by: Lee, Sanwoo, et al.
Published: (2025)

IIET: Efficient Numerical Transformer via Implicit Iterative Euler Method
by: Liu, Xinyu, et al.
Published: (2025)

A Survey on LLM Mid-Training
by: Tu, Chengying, et al.
Published: (2025)

BAPO: Boundary-Aware Policy Optimization for Reliable Agentic Search
by: Liu, Shiyu, et al.
Published: (2026)

Aligning with Human Values to Enhance Interaction: An eHMI-Mediated Lane-Changing Negotiation Strategy Using Bayesian Inference
by: Peng, Boyao, et al.
Published: (2025)

What Makes Quantization for Large Language Models Hard? An Empirical Study from the Lens of Perturbation
by: Gong, Zhuocheng, et al.
Published: (2024)

Resolving Memorization in Empirical Diffusion Model for Manifold Data in High-Dimensional Spaces
by: Lyu, Yang, et al.
Published: (2025)

Deep Learning based Quasi-consciousness Training for Robot Intelligent Model
by: Li, Yuchun, et al.
Published: (2025)

APP: Adaptive Prototypical Pseudo-Labeling for Few-shot OOD Detection
by: Wang, Pei, et al.
Published: (2023)

Unlocking Implicit Experience: Synthesizing Tool-Use Trajectories from Text
by: Xu, Zhihao, et al.
Published: (2026)

Multi-Objective and Mixed-Reward Reinforcement Learning via Reward-Decorrelated Policy Optimization
by: Bai, Yang, et al.
Published: (2026)

Forgetting Curve: A Reliable Method for Evaluating Memorization Capability for Long-context Models
by: Liu, Xinyu, et al.
Published: (2024)

NeedleInATable: Exploring Long-Context Capability of Large Language Models towards Long-Structured Tables
by: Wang, Lanrui, et al.
Published: (2025)

SEAS: Self-Evolving Adversarial Safety Optimization for Large Language Models
by: Diao, Muxi, et al.
Published: (2024)

LinkQA: Synthesizing Diverse QA from Multiple Seeds Strongly Linked by Knowledge Points
by: Zhang, Xuemiao, et al.
Published: (2025)

A Preliminary Study on the Promises and Challenges of Native Top-$k$ Sparse Attention
by: Xiu, Di, et al.
Published: (2025)

NDP: Next Distribution Prediction as a More Broad Target
by: Ruan, Junhao, et al.
Published: (2024)

Refactoring Deep Learning Code: A Study of Practices and Unsatisfied Tool Needs
by: Wang, Siqi, et al.
Published: (2024)

GIFT: Reconciling Post-Training Objectives via Finite-Temperature Gibbs Initialization
by: Zhao, Zhengyang, et al.
Published: (2026)

Rethinking the Sampling Criteria in Reinforcement Learning for LLM Reasoning: A Competence-Difficulty Alignment Perspective
by: Kong, Deyang, et al.
Published: (2025)

Preference Curriculum: LLMs Should Always Be Pretrained on Their Preferred Data
by: Zhang, Xuemiao, et al.
Published: (2025)

Parallel Decoding via Hidden Transfer for Lossless Large Language Model Acceleration
by: Wu, Pengfei, et al.
Published: (2024)

FIRP: Faster LLM inference via future intermediate representation prediction
by: Wu, Pengfei, et al.
Published: (2024)