Image de couverture de livre

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Schaeffer, Rylan, Schoelkopf, Hailey, Miranda, Brando, Mukobi, Gabriel, Madan, Varun, Ibrahim, Adam, Bradley, Herbie, Biderman, Stella, Koyejo, Sanmi
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Artificial Intelligence Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Pretraining Scaling Laws for Generative Evaluations of Language Models
par: Schaeffer, Rylan, et autres
Publié: (2025)

In-Context Learning of Energy Functions
par: Schaeffer, Rylan, et autres
Publié: (2024)

ZIP-FIT: Embedding-Free Data Selection via Compression-Based Alignment
par: Obbad, Elyas, et autres
Publié: (2024)

Beyond Scale: The Diversity Coefficient as a Data Quality Metric for Variability in Natural Language Data
par: Miranda, Brando, et autres
Publié: (2023)

Evaluating the Robustness of Chinchilla Compute-Optimal Scaling
par: Schaeffer, Rylan, et autres
Publié: (2025)

Quantifying the Effect of Test Set Contamination on Generative Evaluations
par: Schaeffer, Rylan, et autres
Publié: (2026)

Understanding Adversarial Transfer: Why Representation-Space Attacks Fail Where Data-Space Attacks Succeed
par: Gupta, Isha, et autres
Publié: (2025)

Position: Model Collapse Does Not Mean What You Think
par: Schaeffer, Rylan, et autres
Publié: (2025)

Quantifying the Importance of Data Alignment in Downstream Model Performance
par: Chawla, Krrish, et autres
Publié: (2025)

Consensus is Not Verification: Why Crowd Wisdom Strategies Fail for LLM Truthfulness
par: Denisov-Blanch, Yegor, et autres
Publié: (2026)

What Causes Polysemanticity? An Alternative Origin Story of Mixed Selectivity from Incidental Causes
par: Lecomte, Victor, et autres
Publié: (2023)

Suppressing Pink Elephants with Direct Principle Feedback
par: Castricato, Louis, et autres
Publié: (2024)

Pantograph: A Machine-to-Machine Interaction Interface for Advanced Theorem Proving, High Level Reasoning, and Data Extraction in Lean 4
par: Aniva, Leni, et autres
Publié: (2024)

Efficient Prediction of Pass@k Scaling in Large Language Models
par: Kazdan, Joshua, et autres
Publié: (2025)

Is Pre-training Truly Better Than Meta-Learning?
par: Miranda, Brando, et autres
Publié: (2023)

Investigating Data Contamination for Pre-training Language Models
par: Jiang, Minhao, et autres
Publié: (2024)

Collapse or Thrive? Perils and Promises of Synthetic Data in a Self-Generating World
par: Kazdan, Joshua, et autres
Publié: (2024)

No, of Course I Can! Deeper Fine-Tuning Attacks That Bypass Token-Level Safety Mechanisms
par: Kazdan, Joshua, et autres
Publié: (2025)

Reasons to Doubt the Impact of AI Risk Evaluations
par: Mukobi, Gabriel
Publié: (2024)

Quantifying Variance in Evaluation Benchmarks
par: Madaan, Lovish, et autres
Publié: (2024)

PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
par: van der Wal, Oskar, et autres
Publié: (2025)

Scaling Laws for Downstream Task Performance of Large Language Models
par: Isik, Berivan, et autres
Publié: (2024)

Why Do Safety Guardrails Degrade Across Languages?
par: Zhang, Max, et autres
Publié: (2026)

Scale Dependent Data Duplication
par: Kazdan, Joshua, et autres
Publié: (2026)

Lean-ing on Quality: How High-Quality Data Beats Diverse Multilingual Data in AutoFormalization
par: Chan, Willy, et autres
Publié: (2025)

Causally Inspired Regularization Enables Domain General Representations
par: Salaudeen, Olawale, et autres
Publié: (2024)

Let's Measure Information Step-by-Step: AI-Based Evaluation Beyond Vibes
par: Robertson, Zachary, et autres
Publié: (2025)

CURE: Cultural Understanding and Reasoning Evaluation - A Framework for "Thick" Culture Alignment Evaluation in LLMs
par: Vo, Truong, et autres
Publié: (2025)

An Elusive Ideal—Why a Pragmatic “Home Time” Quality Measure Remains Hard to Define
par: Kyra O'Brien, et autres
Publié: (2025)

Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Track
par: Schaeffer, Rylan, et autres
Publié: (2025)

Failures to Find Transferable Image Jailbreaks Between Vision-Language Models
par: Schaeffer, Rylan, et autres
Publié: (2024)

In-Situ Behavioral Evaluation for LLM Fairness, Not Standardized-Test Scores
par: Tang, Zeyu, et autres
Publié: (2026)

Best-of-N Jailbreaking
par: Hughes, John, et autres
Publié: (2024)

How Do Large Language Monkeys Get Their Power (Laws)?
par: Schaeffer, Rylan, et autres
Publié: (2025)

Putnam-AXIOM: A Functional and Static Benchmark for Measuring Higher Level Mathematical Reasoning in LLMs
par: Gulati, Aryan, et autres
Publié: (2025)

Uncovering Latent Memories: Assessing Data Leakage and Memorization Patterns in Frontier AI Models
par: Duan, Sunny, et autres
Publié: (2024)

A Framework for Objective-Driven Dynamical Stochastic Fields
par: Zhang, Yibo Jacky, et autres
Publié: (2025)

Llemma: An Open Language Model For Mathematics
par: Azerbayev, Zhangir, et autres
Publié: (2023)

Discovering Implicit Large Language Model Alignment Objectives
par: Chen, Edward, et autres
Publié: (2026)

High-Dimensional Markov-switching Ordinary Differential Processes
par: Tsai, Katherine, et autres
Publié: (2024)