Pith. sign in

REVIEW 3 cited by

Kangaroo: Lossless Self-Speculative Decoding via Double Early Exiting

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2404.18911 v1 pith:XZJWWCMB submitted 2024-04-29 cs.CL cs.LG

classification cs.CLcs.LG
keywords modelkangaroodecodingearlyexitingsub-networktokenacceptance
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
abstract

Speculative decoding has demonstrated its effectiveness in accelerating the inference of large language models while maintaining a consistent sampling distribution. However, the conventional approach of training a separate draft model to achieve a satisfactory token acceptance rate can be costly. Drawing inspiration from early exiting, we propose a novel self-speculative decoding framework \emph{Kangaroo}, which uses a fixed shallow sub-network as a self-draft model, with the remaining layers serving as the larger target model. We train a lightweight and efficient adapter module on top of the sub-network to bridge the gap between the sub-network and the full model's representation ability. It is noteworthy that the inference latency of the self-draft model may no longer be negligible compared to the large model, necessitating strategies to increase the token acceptance rate while minimizing the drafting steps of the small model. To address this challenge, we introduce an additional early exiting mechanism for generating draft tokens. Specifically, we halt the small model's subsequent prediction during the drafting phase once the confidence level for the current token falls below a certain threshold. Extensive experiments on the Spec-Bench demonstrate the effectiveness of Kangaroo. Under single-sequence verification, Kangaroo achieves speedups up to $1.68\times$ on Spec-Bench, outperforming Medusa-1 with 88.7\% fewer additional parameters (67M compared to 591M). The code for Kangaroo is available at https://github.com/Equationliu/Kangaroo.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 3 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Approximate Speculative Decoding

    cs.LG 2026-08 conditional novelty 6.0 of 10

    A training-free verifier, ASD, allows a bounded number of low-regret draft-target mismatches during speculative decoding and reuses the target-greedy suffix, yielding 3 to 15 percent throughput gains with mostly small...

  2. AdaDecode: Accelerating LLM Decoding with Adaptive Layer Parallelism

    cs.CL 2025-06 conditional novelty 6.0 of 10

    AdaDecode speeds up LLM generation by predicting tokens at early layers when confidence is high, running the skipped layers in parallel, and verifying the output exactly matches standard decoding.

  3. CLaSp: In-Context Layer Skip for Self-Speculative Decoding

    cs.CL 2025-05 conditional novelty 6.0 of 10

    A training-free, context-adaptive layer-skipping method for self-speculative decoding that reports roughly 1.1x to 1.8x speedups on LLaMA models while preserving output distribution.

Pith tools