AINeutralarXiv – CS AI · 7h ago6/10
🧠
Soft-NBCE: Entropy-Weighted Chunk Fusion for Long-Context
Researchers introduce Soft-NBCE, an improved method for processing ultra-long text contexts in large language models by replacing discrete chunk selection with weighted chunk fusion. The approach demonstrates measurable improvements on multi-hop reasoning tasks while maintaining efficient memory usage, addressing a critical bottleneck in LLM inference.