1 citations · 1 across the 8 of their papers we have counts for
8 papers
TailSieve: Partial-Rollout-Guided Tail Routing for LLM Rollouts
Tianqi Xu, Lu Lv, Haoyang Huang +15
Large-scale rollouts have become a core component of modern LLM systems, spanning reinforcement learning (RL) post-training, on-policy distillation (OPD), and sampling-heavy evalua…
OmniDelta: Skill-Driven Budget Allocation for Token Compression in OmniLLMs
Haoyang Huang, Wenjie Huang, Tianqi Xu +14
Emerging Omni-modal Large Language Models (OmniLLMs) enable unified understanding of text, audio, and video, but their long audio-video token sequences introduce substantial memory…
When Good Enough Is Optimal: Multiplication-Only Matrix Inversion Approximation for Quantized Gated DeltaNet
Luoming Zhang, Yuwei Ren, Kui Zhang +7
Matrix inversion in chunk-wise parallel linear attention is a major bottleneck for long-context modeling, particularly on NPUs, where forward-substitution-based methods exhibit lim…
Draft Less, Retrieve More: Hybrid Tree Construction for Speculative Decoding
Yuhao Shen, Tianyu Liu, Xinyi Hu +9
Speculative decoding (SD) accelerates large language model inference by leveraging a draft-then-verify paradigm. To maximize the acceptance rate, recent methods construct expansive…
When Hidden States Drift: Can KV Caches Rescue Long-Range Speculative Decoding?
Tianyu Liu, Yuhao Shen, Xinyi Hu +8
Speculative decoding accelerates LLM inference, but SOTA hidden-state-based drafters suffer from long-range decay: draft accuracy degrades as the speculative step increases. Existi…
TALON: Confidence-Aware Speculative Decoding with Adaptive Token Trees
Tianyu Liu, Qitan Lv, Yuhao Shen +2
Speculative decoding (SD) has become a standard technique for accelerating LLM inference without sacrificing output quality. Recent advances in speculative decoding have shifted fr…
Double: Breaking the Acceleration Limit via Double Retrieval Speculative Parallelism
Yuhao Shen, Tianyu Liu, Junyi Shen +4
Parallel Speculative Decoding (PSD) accelerates traditional Speculative Decoding (SD) by overlapping draft generation with verification. However, it remains hampered by two fundame…
SpecBranch: Speculative Decoding via Hybrid Drafting and Rollback-Aware Branch Parallelism
Yuhao Shen, Junyi Shen, Quan Kong +3
Recently, speculative decoding (SD) has emerged as a promising technique to accelerate LLM inference by employing a small draft model to propose draft tokens in advance, and valida…