SpecBlock achieves 8-19% higher speedup than EAGLE-3 in LLM speculative decoding by using repeated block expansions with hidden-state inheritance, a dynamic rank head, and a valid-prefix training mask.
Stanford alpaca: An instruction-following llama model
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
cs.CL 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
ReAD applies a contextual bandit to allocate fixed-token distillation budget across interdependent LLM capabilities, yielding higher task utility and fewer negative spillovers than standard methods.
LoPT achieves competitive task performance in LLM post-training by limiting task gradients to the upper model half and training the lower half with local feature reconstruction.
citing papers explorer
-
SpecBlock: Block-Iterative Speculative Decoding with Dynamic Tree Drafting
SpecBlock achieves 8-19% higher speedup than EAGLE-3 in LLM speculative decoding by using repeated block expansions with hidden-state inheritance, a dynamic rank head, and a valid-prefix training mask.
-
ReAD: Reinforcement-Guided Capability Distillation for Large Language Models
ReAD applies a contextual bandit to allocate fixed-token distillation budget across interdependent LLM capabilities, yielding higher task utility and fewer negative spillovers than standard methods.
-
Rethinking Local Learning: A Cheaper and Faster Recipe for LLM Post-Training
LoPT achieves competitive task performance in LLM post-training by limiting task gradients to the upper model half and training the lower half with local feature reconstruction.