Copertina

Unveiling Downstream Performance Scaling of LLMs: A Clustering-Based Perspective

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Xu, Chengyin, Chen, Kaiyuan, Li, Xiao, Shen, Ke, Li, Chenggang
Natura:	Preprint
Pubblicazione:	2025
Soggetti:	Computation and Language Artificial Intelligence Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Scaling Laws for Predicting Downstream Performance in LLMs
di: Chen, Yangyi, et al.
Pubblicazione: (2024)

Quantifying the Importance of Data Alignment in Downstream Model Performance
di: Chawla, Krrish, et al.
Pubblicazione: (2025)

Revisiting the Scaling Properties of Downstream Metrics in Large Language Model Training
di: Krajewski, Jakub, et al.
Pubblicazione: (2025)

Task-Informed Anti-Curriculum by Masking Improves Downstream Performance on Text
di: Jarca, Andrei, et al.
Pubblicazione: (2025)

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?
di: Schaeffer, Rylan, et al.
Pubblicazione: (2024)

KcMF: A Knowledge-compliant Framework for Schema and Entity Matching with Fine-tuning-free LLMs
di: Xu, Yongqin, et al.
Pubblicazione: (2024)

Understanding and Mitigating Bias Inheritance in LLM-based Data Augmentation on Downstream Tasks
di: Li, Miaomiao, et al.
Pubblicazione: (2025)

ClusterUCB: Efficient Gradient-Based Data Selection for Targeted Fine-Tuning of LLMs
di: Wang, Zige, et al.
Pubblicazione: (2025)

Principled RL for Diffusion LLMs Emerges from a Sequence-Level Perspective
di: Ou, Jingyang, et al.
Pubblicazione: (2025)

Are the Values of LLMs Structurally Aligned with Humans? A Causal Perspective
di: Kang, Yipeng, et al.
Pubblicazione: (2024)

Product of Experts with LLMs: Boosting Performance on ARC Is a Matter of Perspective
di: Franzen, Daniel, et al.
Pubblicazione: (2025)

What Happened in LLMs Layers when Trained for Fast vs. Slow Thinking: A Gradient Perspective
di: Li, Ming, et al.
Pubblicazione: (2024)

How Data Inter-connectivity Shapes LLMs Unlearning: A Structural Unlearning Perspective
di: Qiu, Xinchi, et al.
Pubblicazione: (2024)

A Decomposition Perspective to Long-context Reasoning for LLMs
di: Xiao, Yanling, et al.
Pubblicazione: (2026)

Learning to Rewrite Prompts for Bootstrapping LLMs on Downstream Tasks
di: Zhou, Qinhao, et al.
Pubblicazione: (2025)

RAG-Instruct: Boosting LLMs with Diverse Retrieval-Augmented Instructions
di: Liu, Wanlong, et al.
Pubblicazione: (2024)

Rethinking the Role of Prompting Strategies in LLM Test-Time Scaling: A Perspective of Probability Theory
di: Liu, Yexiang, et al.
Pubblicazione: (2025)

Tactic: Adaptive Sparse Attention with Clustering and Distribution Fitting for Long-Context LLMs
di: Zhu, Kan, et al.
Pubblicazione: (2025)

Diversity as a Reward: Fine-Tuning LLMs on a Mixture of Domain-Undetermined Data
di: Ling, Zhenqing, et al.
Pubblicazione: (2025)

Configurable Foundation Models: Building LLMs from a Modular Perspective
di: Xiao, Chaojun, et al.
Pubblicazione: (2024)

Reasoning-Based Refinement of Unsupervised Text Clusters with LLMs
di: Islam, Tunazzina
Pubblicazione: (2026)

FBI-LLM: Scaling Up Fully Binarized LLMs from Scratch via Autoregressive Distillation
di: Ma, Liqun, et al.
Pubblicazione: (2024)

DiLoCoX: A Low-Communication Large-Scale Training Framework for Decentralized Cluster
di: Qi, Ji, et al.
Pubblicazione: (2025)

MarkovScale: Towards Optimal Sequential Scaling at Inference Time
di: Wang, Youkang, et al.
Pubblicazione: (2026)

AutoScale: Scale-Aware Data Mixing for Pre-Training LLMs
di: Kang, Feiyang, et al.
Pubblicazione: (2024)

Cer-Eval: Certifiable and Cost-Efficient Evaluation Framework for LLMs
di: Wang, Ganghua, et al.
Pubblicazione: (2025)

Multitask Mayhem: Unveiling and Mitigating Safety Gaps in LLMs Fine-tuning
di: Jan, Essa, et al.
Pubblicazione: (2024)

Self-Distillation as a Performance Recovery Mechanism for LLMs: Counteracting Compression and Catastrophic Forgetting
di: Liu, Chi, et al.
Pubblicazione: (2026)

Mission Impossible: A Statistical Perspective on Jailbreaking LLMs
di: Su, Jingtong, et al.
Pubblicazione: (2024)

Robo2VLM: Visual Question Answering from Large-Scale In-the-Wild Robot Manipulation Datasets
di: Chen, Kaiyuan, et al.
Pubblicazione: (2025)

Instruction Learning Paradigms: A Dual Perspective on White-box and Black-box LLMs
di: Ren, Yanwei, et al.
Pubblicazione: (2025)

Graph Learning in the Era of LLMs: A Survey from the Perspective of Data, Models, and Tasks
di: Li, Xunkai, et al.
Pubblicazione: (2024)

TDBench: A Benchmark for Top-Down Image Understanding with Reliability Analysis of Vision-Language Models
di: Hou, Kaiyuan, et al.
Pubblicazione: (2025)

VocabTailor: Dynamic Vocabulary Selection for Downstream Tasks in Small Language Models
di: Zhang, Hanling, et al.
Pubblicazione: (2025)

ALPINE: Unveiling the Planning Capability of Autoregressive Learning in Language Models
di: Wang, Siwei, et al.
Pubblicazione: (2024)

Unveiling and Addressing Pseudo Forgetting in Large Language Models
di: Sun, Huashan, et al.
Pubblicazione: (2024)

Can LLMs Speak For Diverse People? Tuning LLMs via Debate to Generate Controllable Controversial Statements
di: Li, Ming, et al.
Pubblicazione: (2024)

Benefits and Pitfalls of Reinforcement Learning for Language Model Planning: A Theoretical Perspective
di: Wang, Siwei, et al.
Pubblicazione: (2025)

Benchmarking for Domain-Specific LLMs: A Case Study on Academia and Beyond
di: Chen, Rubing, et al.
Pubblicazione: (2025)

DLO: Dynamic Layer Operation for Efficient Vertical Scaling of LLMs
di: Tan, Zhen, et al.
Pubblicazione: (2024)