Compression Represents Intelligence Linearly
Paper
•
2404.09937
•
Published
•
27
LM Capabilities and Scaling
-
-
MiniCPM: Unveiling the Potential of Small Language Models with Scalable Training Strategies
Paper • 2404.06395 • Published • 22 -
Long-context LLMs Struggle with Long In-context Learning
Paper • 2404.02060 • Published • 37 -
Are large language models superhuman chemists?
Paper • 2404.01475 • Published • 19 -
FlowMind: Automatic Workflow Generation with LLMs
Paper • 2404.13050 • Published • 34 -
Capabilities of Gemini Models in Medicine
Paper • 2404.18416 • Published • 24 -
Imp: Highly Capable Large Multimodal Models for Mobile Devices
Paper • 2405.12107 • Published • 29 -
On the Planning Abilities of Large Language Models -- A Critical Investigation
Paper • 2305.15771 • Published • 1 -
Test of Time: A Benchmark for Evaluating LLMs on Temporal Reasoning
Paper • 2406.09170 • Published • 27 -
MuirBench: A Comprehensive Benchmark for Robust Multi-image Understanding
Paper • 2406.09411 • Published • 20 -
Accessing GPT-4 level Mathematical Olympiad Solutions via Monte Carlo Tree Self-refine with LLaMa-3 8B
Paper • 2406.07394 • Published • 28 -
GEB-1.3B: Open Lightweight Large Language Model
Paper • 2406.09900 • Published • 21 -
Mixture of A Million Experts
Paper • 2407.04153 • Published • 5 -
Physics of Language Models: Part 3.3, Knowledge Capacity Scaling Laws
Paper • 2404.05405 • Published • 10 -
Mutual Reasoning Makes Smaller LLMs Stronger Problem-Solvers
Paper • 2408.06195 • Published • 71 -
Attention Heads of Large Language Models: A Survey
Paper • 2409.03752 • Published • 89 -
HelloBench: Evaluating Long Text Generation Capabilities of Large Language Models
Paper • 2409.16191 • Published • 42 -
Making Text Embedders Few-Shot Learners
Paper • 2409.15700 • Published • 30 -
Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data
Paper • 2406.14546 • Published • 2 -
Are Your LLMs Capable of Stable Reasoning?
Paper • 2412.13147 • Published • 94 -
CodeElo: Benchmarking Competition-level Code Generation of LLMs with Human-comparable Elo Ratings
Paper • 2501.01257 • Published • 52 -
ProgCo: Program Helps Self-Correction of Large Language Models
Paper • 2501.01264 • Published • 26 -
Densing Law of LLMs
Paper • 2412.04315 • Published • 19 -
Low-Bit Quantization Favors Undertrained LLMs: Scaling Laws for Quantized LLMs with 100T Training Tokens
Paper • 2411.17691 • Published • 13 -
PokerBench: Training Large Language Models to become Professional Poker Players
Paper • 2501.08328 • Published • 17 -
Do generative video models learn physical principles from watching videos?
Paper • 2501.09038 • Published • 33 -
Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models
Paper • 2501.12370 • Published • 11 -
Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
Paper • 2501.16975 • Published • 28 -
Large Language Models Think Too Fast To Explore Effectively
Paper • 2501.18009 • Published • 23 -
The Stochastic Parrot on LLM's Shoulder: A Summative Assessment of Physical Concept Understanding
Paper • 2502.08946 • Published • 192 -
Scaling Embedding Layers in Language Models
Paper • 2502.01637 • Published • 24 -
Great Models Think Alike and this Undermines AI Oversight
Paper • 2502.04313 • Published • 33 -
Scaling Pre-training to One Hundred Billion Data for Vision Language Models
Paper • 2502.07617 • Published • 29 -
Gemstones: A Model Suite for Multi-Faceted Scaling Laws
Paper • 2502.06857 • Published • 25 -
Distillation Scaling Laws
Paper • 2502.08606 • Published • 46 -
NoLiMa: Long-Context Evaluation Beyond Literal Matching
Paper • 2502.05167 • Published • 15 -
Cramming 1568 Tokens into a Single Vector and Back Again: Exploring the Limits of Embedding Space Capacity
Paper • 2502.13063 • Published • 68 -
Unveiling Downstream Performance Scaling of LLMs: A Clustering-Based Perspective
Paper • 2502.17262 • Published • 20