Cassandra is a self-speculative decoding system that builds a draft model via fine-grained data selection and optimized pruning/mantissa truncation, achieving up to 2.41x speedup over BF16 and 1.81x more tokens than Eagle-3 on Llama 3 8B without training.
Quantspec: Self-speculative decoding with hierarchical quantized kv cache,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AR 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Cassandra: Enabling Reasoning LLMs at Edge via Self-Speculative Decoding
Cassandra is a self-speculative decoding system that builds a draft model via fine-grained data selection and optimized pruning/mantissa truncation, achieving up to 2.41x speedup over BF16 and 1.81x more tokens than Eagle-3 on Llama 3 8B without training.