Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning
Large language models often rely on Chain-of-Thought (CoT) reasoning to solve complex tasks, but verbose reasoning traces introduce substantial inference overhead. CoT compression shortens generation, yet aggressive compression may disrupt logical coherence and degrade performance. We formalize this trade-off as the \textit{Context-Generation Substitution Law}, where explicit reasoning context substitutes for part of decode-time generation. Based on this principle, we propose \textit{Memory-Augmented Compression}, a training-free framework that constructs reusable reasoning memories from histo
Lineage graph
Paper → model → repo connections mined from source citations (Tier-1 exact match).
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- FuzzyOverlapping authors or contributors · 62%modular/modular →
“Shared author/contributor keys: liu”
- LinkedLinked via arxiv author · 85%Simeng Zhang →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Yilong Chen →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Wenyuan Zhang →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Zhenyu Zhang →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Yao Cheng Li →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Junyuan Shang →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”
- LinkedLinked via arxiv author · 85%Tingwen Liu →
“Memory Augmentation Unlocks Efficient Chain-of-Thought Reasoning”