作者位次 #3/4ACL 2026Findings· Findings of ACL
TELL - TALE : Task Efficient LLM s with Task Aware Layer Elimination
Omar Naim · Krish Sharma · Niyar R Barman · Nicholas Asher
摘要
Large Language Models (LLMs) typically come with a fixed architecture, despite growing evidence that not all layers contribute equally to every downstream task. We introduce TALE (Task-Aware Layer Elimination), an inference-time method that improves task performance by selectively removing layers...