pith. sign in

Consequently, it triggers the out-of-memory exception at an early stage (128K tokens)

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.CL 1

years

2026 1

verdicts

UNVERDICTED 1

representative citing papers

Stacked from One: Multi-Scale Self-Injection for Context Window Extension

cs.CL · 2026-03-05 · unverdicted · novelty 6.0

SharedLLM stacks two copies of a short-context LLM so the lower one compresses context into query-aware multi-grained tokens that are injected only at the lowest layers of the upper one, enabling generalization from 8K training to 128K+ inputs.

citing papers explorer

Showing 1 of 1 citing paper.

  • Stacked from One: Multi-Scale Self-Injection for Context Window Extension cs.CL · 2026-03-05 · unverdicted · none · ref 37

    SharedLLM stacks two copies of a short-context LLM so the lower one compresses context into query-aware multi-grained tokens that are injected only at the lowest layers of the upper one, enabling generalization from 8K training to 128K+ inputs.