Image de couverture de livre

Learning from the Best, Differently: A Diversity-Driven Rethinking on Data Selection

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	He, Hongyi, Liu, Xiao, Lin, Zhenghao, Tang, Mingni, Cheng, Yi, Wang, Jintao, Li, Wenjie, Cheng, Peng, Gong, Yeyun
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Improving Data and Reward Design for Scientific Reasoning in Large Language Models
par: Chen, Zijie, et autres
Publié: (2026)

Enhancing Large Language Model Performance with Gradient-Based Parameter Selection
par: Li, Haoling, et autres
Publié: (2024)

Data Mixing Agent: Learning to Re-weight Domains for Continual Pre-training
par: Yang, Kailai, et autres
Publié: (2025)

Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoning
par: Huang, Yiming, et autres
Publié: (2024)

Beyond Length: Quantifying Long-Range Information for Long-Context LLM Pretraining Data
par: Deng, Haoran, et autres
Publié: (2025)

Overcoming Vocabulary Mismatch: Vocabulary-agnostic Teacher Guided Language Modeling
par: Shin, Haebin, et autres
Publié: (2025)

Integrative Decoding: Improve Factuality via Implicit Self-consistency
par: Cheng, Yi, et autres
Publié: (2024)

Ensuring Safe and High-Quality Outputs: A Guideline Library Approach for Language Models
par: Luo, Yi, et autres
Publié: (2024)

Competition-Level Problems are Effective LLM Evaluators
par: Huang, Yiming, et autres
Publié: (2023)

Rho-1: Not All Tokens Are What You Need
par: Lin, Zhenghao, et autres
Publié: (2024)

DSGBench: A Diverse Strategic Game Benchmark for Evaluating LLM-based Agents in Complex Decision-Making Environments
par: Tang, Wenjie, et autres
Publié: (2025)

Sigma-MoE-Tiny Technical Report
par: Hu, Qingguo, et autres
Publié: (2025)

Rethinking Data Selection at Scale: Random Selection is Almost All You Need
par: Xia, Tingyu, et autres
Publié: (2024)

LayerNorm Induces Recency Bias in Transformer Decoders
par: Kim, Junu, et autres
Publié: (2025)

Enhancing Chain-of-Thoughts Prompting with Iterative Bootstrapping in Large Language Models
par: Sun, Jiashuo, et autres
Publié: (2023)

Toolink: Linking Toolkit Creation and Using through Chain-of-Solving on Open-Source Model
par: Qian, Cheng, et autres
Publié: (2023)

Process-based Self-Rewarding Language Models
par: Zhang, Shimao, et autres
Publié: (2025)

From Intention To Implementation: Automating Biomedical Research via LLMs
par: Luo, Yi, et autres
Publié: (2024)

DynamixSFT: Dynamic Mixture Optimization of Instruction Tuning Collections
par: Shin, Haebin, et autres
Publié: (2025)

Learning Generative Selection for Best-of-N
par: Toshniwal, Shubham, et autres
Publié: (2026)

How Does Alignment Enhance LLMs' Multilingual Capabilities? A Language Neurons Perspective
par: Zhang, Shimao, et autres
Publié: (2025)

The Strongest Teacher Is Not Always the Best Teacher: Student-Centric Answer Selection
par: Hu, Zhengyu, et autres
Publié: (2026)

Rethinking Machine Unlearning in Image Generation Models
par: Liu, Renyang, et autres
Publié: (2025)

TACOS: Open Tagging and Comparative Scoring for Instruction Fine-Tuning Data Selection
par: He, Xixiang, et autres
Publié: (2025)

Velocitune: A Velocity-based Dynamic Domain Reweighting Method for Continual Pre-training
par: Luo, Zheheng, et autres
Publié: (2024)

Curriculum Learning with Quality-Driven Data Selection
par: Wu, Biao, et autres
Publié: (2024)

Rethinking Diverse Human Preference Learning through Principal Component Analysis
par: Luo, Feng, et autres
Publié: (2025)

The Best Instruction-Tuning Data are Those That Fit
par: Zhang, Dylan, et autres
Publié: (2025)

HiCaM: A Hierarchical-Causal Modification Framework for Long-Form Text Modification
par: Shi, Yuntao, et autres
Publié: (2025)

Generative Prompt Internalization
par: Shin, Haebin, et autres
Publié: (2024)

Adapting LLM Agents with Universal Feedback in Communication
par: Wang, Kuan, et autres
Publié: (2023)

Selective Expert Guidance for Effective and Diverse Exploration in Reinforcement Learning of LLMs
par: Jiang, Zishang, et autres
Publié: (2025)

Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities
par: Dai, Qirun, et autres
Publié: (2025)

Medical Dialogue Generation via Intuitive-then-Analytical Differential Diagnosis
par: Xu, Kaishuai, et autres
Publié: (2024)

Reasoning Like a Doctor: Improving Medical Dialogue Systems via Diagnostic Reasoning Process Alignment
par: Xu, Kaishuai, et autres
Publié: (2024)

Ensembling Large Language Models with Process Reward-Guided Tree Search for Better Complex Reasoning
par: Park, Sungjin, et autres
Publié: (2024)

Diversify and Conquer: Diversity-Centric Data Selection with Iterative Refinement
par: Yu, Simon, et autres
Publié: (2024)

Scoring, Reasoning, and Selecting the Best! Ensembling Large Language Models via a Peer-Review Process
par: Chen, Zhijun, et autres
Publié: (2025)

Pull Requests as a Training Signal for Repo-Level Code Editing
par: Zhu, Qinglin, et autres
Publié: (2026)

Learning to Align Multi-Faceted Evaluation: A Unified and Robust Framework
par: Xu, Kaishuai, et autres
Publié: (2025)