Reward-Guided Speculative Decoding for Efficient LLM Reasoning
Paper
•
2501.19324
•
Published
•
30
Papers
-
-
s1: Simple test-time scaling
Paper • 2501.19393 • Published • 72 -
Scalable-Softmax Is Superior for Attention
Paper • 2501.19399 • Published • 17 -
The Surprising Agreement Between Convex Optimization Theory and Learning-Rate Scheduling for Large Model Training
Paper • 2501.18965 • Published • 5 -
Trading Inference-Time Compute for Adversarial Robustness
Paper • 2501.18841 • Published • 3 -
ChunkKV: Semantic-Preserving KV Cache Compression for Efficient Long-Context LLM Inference
Paper • 2502.00299 • Published • 1 -
Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs
Paper • 2501.18585 • Published • 46 -
Large Language Models Think Too Fast To Explore Effectively
Paper • 2501.18009 • Published • 22 -
SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
Paper • 2501.17161 • Published • 99 -
Over-Tokenized Transformer: Vocabulary is Generally Worth Scaling
Paper • 2501.16975 • Published • 21 -
Scaling Embedding Layers in Language Models
Paper • 2502.01637 • Published • 11 -
DeepRAG: Thinking to Retrieval Step by Step for Large Language Models
Paper • 2502.01142 • Published • 10 -
PaSa: An LLM Agent for Comprehensive Academic Paper Search
Paper • 2501.10120 • Published • 42 -
MiniMax-01: Scaling Foundation Models with Lightning Attention
Paper • 2501.08313 • Published • 272 -
Tensor Product Attention Is All You Need
Paper • 2501.06425 • Published • 80