Limits of Reliability and Scaling in Language Models
cs.CL, cs.AI, cs.IT, math.IT
Submitted: 2026-05-08
Updated: 2026-09-17
Comments: 41 pages, 2 figures
Code: https://github.com/evaleval/benchmark-saturation
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation
- Bayesian scaling laws for in-context learning
- Program Synthesis with Large Language Models
- Explaining Neural Scaling Laws
- Chinchilla Scaling: A replication attempt
- Large Language Monkeys: Scaling Inference Compute with Repeated Sampling
- Broken Neural Scaling Laws
- Deriving Neural Scaling Laws from the statistics of natural language
- Evaluating Large Language Models Trained on Code
- Training Verifiers to Solve Math Word Problems
- Hierarchical Neural Story Generation
- The Llama 3 Herd of Models
- Understanding Scaling Laws with Statistical and Approximation Theory for Transformer Neural Networks on Intrinsically Low-dimensional Data
- Scaling Laws for Neural Language Models
- Revisiting LLM Reasoning via Information Bottleneck
- On the Fundamental Limits of LLMs at Scale
- NC-Bench: An LLM Benchmark for Evaluating Conversational Competence
- EQ-Bench: An Emotional Intelligence Benchmark for Large Language Models
- Get To The Point: Summarization with Pointer-Generator Networks
- Opening the Black Box of Deep Neural Networks via Information
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering