Strong Model Collapse

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Dohmatob, Elvis, Feng, Yunzhen, Subramonian, Arjun, Kempe, Julia
Natura:	Preprint
Pubblicazione:	2024
Soggetti:	Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

Model Collapse Demystified: The Case of Regression
di: Dohmatob, Elvis, et al.
Pubblicazione: (2024)

Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verification
di: Feng, Yunzhen, et al.
Pubblicazione: (2024)

auto-fpt: Automating Free Probability Theory Calculations for Machine Learning Theory
di: Subramonian, Arjun, et al.
Pubblicazione: (2025)

A Tale of Tails: Model Collapse as a Change of Scaling Laws
di: Dohmatob, Elvis, et al.
Pubblicazione: (2024)

An Effective Theory of Bias Amplification
di: Subramonian, Arjun, et al.
Pubblicazione: (2024)

PILAF: Optimal Human Preference Sampling for Reward Modeling
di: Feng, Yunzhen, et al.
Pubblicazione: (2025)

Egalitarian Gradient Descent: A Simple Approach to Accelerated Grokking
di: Pasand, Ali Saheb, et al.
Pubblicazione: (2025)

Don't Waste Mistakes: Leveraging Negative RL-Groups via Confidence Reweighting
di: Feng, Yunzhen, et al.
Pubblicazione: (2025)

What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of CoT
di: Feng, Yunzhen, et al.
Pubblicazione: (2025)

Efficient Refusal Ablation in LLM through Optimal Transport
di: Nanfack, Geraldin, et al.
Pubblicazione: (2026)

Why Less is More (Sometimes): A Theory of Data Curation
di: Dohmatob, Elvis, et al.
Pubblicazione: (2025)

On the Robustness of Neural Collapse and the Neural Collapse of Robustness
di: Su, Jingtong, et al.
Pubblicazione: (2023)

Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks
di: Feng, Yunzhen, et al.
Pubblicazione: (2024)

Theoretical and Empirical Insights into the Origins of Degree Bias in Graph Neural Networks
di: Subramonian, Arjun, et al.
Pubblicazione: (2024)

Scaling Laws for Associative Memories
di: Cabannes, Vivien, et al.
Pubblicazione: (2023)

Networked Inequality: Preferential Attachment Bias in Graph Neural Network Link Prediction
di: Subramonian, Arjun, et al.
Pubblicazione: (2023)

The Pitfalls of Memorization: When Memorization Hurts Generalization
di: Bayat, Reza, et al.
Pubblicazione: (2024)

Weisfeiler and Leman Go Measurement Modeling: Probing the Validity of the WL Test
di: Subramonian, Arjun, et al.
Pubblicazione: (2023)

Emergent properties with repeated examples
di: Charton, François, et al.
Pubblicazione: (2024)

Deconstructing the Goldilocks Zone of Neural Network Initialization
di: Vysogorets, Artem, et al.
Pubblicazione: (2024)

Outcome-based Exploration for LLM Reasoning
di: Song, Yuda, et al.
Pubblicazione: (2025)

Flavors of Margin: Implicit Bias of Steepest Descent in Homogeneous Neural Networks
di: Tsilivis, Nikolaos, et al.
Pubblicazione: (2024)

The Price of Implicit Bias in Adversarially Robust Generalization
di: Tsilivis, Nikolaos, et al.
Pubblicazione: (2024)

How Reinforcement Learning After Next-Token Prediction Facilitates Learning
di: Tsilivis, Nikolaos, et al.
Pubblicazione: (2025)

Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective
di: Yao, Yunzhen, et al.
Pubblicazione: (2025)

Non-Asymptotic Analysis of Efficiency in Conformalized Regression
di: Yao, Yunzhen, et al.
Pubblicazione: (2025)

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

DRoP: Distributionally Robust Data Pruning
di: Vysogorets, Artem, et al.
Pubblicazione: (2024)

From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers
di: Su, Jingtong, et al.
Pubblicazione: (2025)

Spend Wisely: Maximizing Post-Training Gains in Iterative Synthetic Data Bootstrapping
di: Yang, Pu, et al.
Pubblicazione: (2025)

Improving the Scaling Laws of Synthetic Data with Deliberate Practice
di: Askari-Hemmat, Reyhane, et al.
Pubblicazione: (2025)

Efficient RL Training for LLMs with Experience Replay
di: Arnal, Charles, et al.
Pubblicazione: (2026)

Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability
di: Sundaram, Shobhita, et al.
Pubblicazione: (2026)

Preventing Model Collapse in Gaussian Process Latent Variable Models
di: Li, Ying, et al.
Pubblicazione: (2024)

Embedding Trust: Semantic Isotropy Predicts Nonfactuality in Long-Form Text Generation
di: Bhardwaj, Dhrupad, et al.
Pubblicazione: (2025)

DeLTa: A Decoding Strategy based on Logit Trajectory Prediction Improves Factuality and Reasoning Ability
di: He, Yunzhen, et al.
Pubblicazione: (2025)

On the Geometry of Regularization in Adversarial Training: High-Dimensional Asymptotics and Generalization Bounds
di: Vilucchio, Matteo, et al.
Pubblicazione: (2024)

Stability and Multigroup Fairness in Ranking with Uncertain Predictions
di: Devic, Siddartha, et al.
Pubblicazione: (2024)

Soft Tokens, Hard Truths
di: Butt, Natasha, et al.
Pubblicazione: (2025)

Asymmetric REINFORCE for off-Policy Reinforcement Learning: Balancing positive and negative rewards
di: Arnal, Charles, et al.
Pubblicazione: (2025)