MétaCan
Menu
Back to cohort
Record W4414726848 · doi:10.1145/3737902.3768352

WiP: Efficient Speculative Decoding for AI PCs via Hierarchical N-Gram Retrieval

2025· article· en· W4414726848 on OpenAlexaff
Huajun Bai, Zihao An, Qing Wang, Qi Tian, Dong Li, Emad Barsoum, Jiwu Shu

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicAlgorithms and Data Compression
Canadian institutionsAdvanced Micro Devices (Canada)
Fundersnot available
KeywordsDecoding methodsSpeedupInferenceLook-aheadQuality (philosophy)Face (sociological concept)

Abstract

fetched live from OpenAlex

AI PCs are emerging as a promising platform for on-device LLM inference, but their limited memory and compute resources pose significant challenges to Large Language Model (LLM) inference, which underutilizes available computing resources in decoding. Speculative decoding addresses this by generating multiple draft tokens and verifying them in parallel using the target model, improving efficiency over purely autoregressive methods. However, existing speculative decoding methods struggle on AI PCs: model-based approaches are bottlenecked by non-negligible draft model inference time, while retrieval-based methods face a trade-off between insufficient retrieval quality and excessive memory usage from external datasets. To address this fundamental trade-off, we propose NG+, a work-in-progress retrieval-based LLM inference system optimized for AI PCs. NG+ is designed to deliver retrieval quality of large-scale datasets within constrained memory budgets by employing a hierarchical n-gram caching architecture---storing frequent n-grams in memory and offloading less frequent ones to SSD. In addition, NG+ overlaps CPU-based n-gram retrieval for the next decoding step with iGPU-based verification of the current draft, effectively prefetching from SSD to hide I/O latency. Preliminary evaluations on an AMD Ryzen™ AI MAX+ 395 platform demonstrate NG+'s efficacy, achieving a 1.62x speedup over auto-regressive decoding, outperforming existing baselines.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.006
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.012
Threshold uncertainty score0.024

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0010.006
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0010.001
Bibliometrics0.0010.002
Science and technology studies0.0010.001
Scholarly communication0.0020.004
Open science0.0030.002
Research integrity0.0010.002
Insufficient payload (model declined to judge)0.0060.006

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.012
GPT teacher head0.299
Teacher spread0.287 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations0
Published2025
Admission routes1
Has abstractyes

Explore more

Same topicAlgorithms and Data CompressionFrench-language works237,207