Cover Image

Accelerating RL Post-Training Rollouts via System-Integrated Speculative Decoding

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Iso, Hayate, Mitra, Tiyasa, Mondal, Sudipta, Shafipour, Rasoul, Elango, Venmugil, Kong, Terry, Huang, Yuki, Na, Seonjin, Putterman, Izzy, Chislett, Benjamin, Ashkenazi, Maor, Guman, Joseph, Shen, Gerald, Konuk, Tugrul, Aithal, Ashwath, Borkar, Ritika, Zilberstein, Ran, Rouhani, Bita
Format:	Preprint
Published:	2026
Subjects:	Machine Learning Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

SPEED-Bench: A Unified and Diverse Benchmark for Speculative Decoding
by: Abramovich, Talor, et al.
Published: (2026)

LatentMoE: Toward Optimal Accuracy per FLOP and Parameter in Mixture of Experts
by: Elango, Venmugil, et al.
Published: (2026)

ATTENTION2D: Communication Efficient Distributed Self-Attention Mechanism
by: Elango, Venmugil
Published: (2025)

PaSE: Parallelization Strategies for Efficient DNN Training
by: Elango, Venmugil
Published: (2024)

AutoTemplate: A Simple Recipe for Lexically Constrained Text Generation
by: Iso, Hayate
Published: (2022)

Helix Parallelism: Rethinking Sharding Strategies for Interactive Multi-Million-Token LLM Decoding
by: Bhatia, Nidhi, et al.
Published: (2025)

Elastic least‐squares reverse time migration from topography through anisotropic tensorial elastodynamics
by: Tugrul Konuk, et al.
Published: (2024)

AmbigNLG: Addressing Task Ambiguity in Instruction for NLG
by: Niwa, Ayana, et al.
Published: (2024)

Beyond the Buzz: A Pragmatic Take on Inference Disaggregation
by: Mitra, Tiyasa, et al.
Published: (2025)

Towards Croppable Implicit Neural Representations
by: Ashkenazi, Maor, et al.
Published: (2024)

Tied-Lora: Enhancing parameter efficiency of LoRA with weight tying
by: Renduchintala, Adithya, et al.
Published: (2023)

Noisy Pairing and Partial Supervision for Stylized Opinion Summarization
by: Iso, Hayate, et al.
Published: (2022)

Holistic Reasoning with Long-Context LMs: A Benchmark for Database Operations on Massive Textual Data
by: Maekawa, Seiji, et al.
Published: (2024)

The Rarity Blind Spot: A Framework for Evaluating Statistical Reasoning in LLMs
by: Maekawa, Seiji, et al.
Published: (2025)

XATU: A Fine-grained Instruction-based Benchmark for Explainable Text Updates
by: Zhang, Haopeng, et al.
Published: (2023)

Evaluating Bias in LLMs for Job-Resume Matching: Gender, Race, and Education
by: Iso, Hayate, et al.
Published: (2025)

Retrieval Helps or Hurts? A Deeper Dive into the Efficacy of Retrieval Augmentation to Language Models
by: Maekawa, Seiji, et al.
Published: (2024)

Key, Value, Compress: A Systematic Exploration of KV Cache Compression Techniques
by: Javidnia, Neusha, et al.
Published: (2025)

Emission Distribution for the quantas of Maxwell-Chern-Simon Gauge Field coupled to External Current
by: Kar, Tiyasa
Published: (2021)

Less is More for Long Document Summary Evaluation by LLMs
by: Wu, Yunshu, et al.
Published: (2023)

Efficient MoE Serving in the Memory-Bound Regime: Balance Activated Experts, Not Tokens
by: Yu, Yanpeng, et al.
Published: (2025)

Zero-Shot Detection of LLM-Generated Code via Approximated Task Conditioning
by: Ashkenazi, Maor, et al.
Published: (2025)

La Psicología y La Psicoterapia en Otros Países "La Psicología Tiene un Largo Pasado Pero una Historia Corta": Turquía, como un ejemplo
by: Emre Konuk
Published: (2011)

From Single to Multi: How LLMs Hallucinate in Multi-Document Summarization
by: Belem, Catarina G., et al.
Published: (2024)

Llama 3 Meets MoE: Efficient Upcycling
by: Vavre, Aditya, et al.
Published: (2024)

Asymptotic expansion of the weighted power variation with second order differences of a stochastic differential equation driven by fBm
by: Yamagishi, Hayate
Published: (2024)

Seat number configuration of the box-ball system, and its relation to the 10-elimination and invariant measures
by: Suda, Hayate
Published: (2023)

Asymptotic expansion of a Hurst index estimator for a stochastic differential equation driven by fBm
by: Yamagishi, Hayate
Published: (2024)

The MacWilliams Identity for Krawtchouk Association Schemes
by: Friedlander, Izzy
Published: (2024)

The MacWilliams Identity for the Hermitian Rank Metric
by: Friedlander, Izzy
Published: (2023)

An Investigation into Emotional Intelligence, Foreign Language Anxiety and Empathy through a Cognitive-Affective Course in an EFL Context
by: Ali Rouhani
Published: (2008)

AI-driven GPT Prompts for Industry Analysis Scholarly Article
by: Aithal, Sreeramana
Published: (2025)

Ethical Leadership in the Light of the Bhagavad Gita: Exploring Dharma-Centered Management Practices for the 21st Century
by: Aithal, Sreeramana
Published: (2025)

Outcome Logic: A Unified Approach to the Metatheory of Program Logics with Branching Effects
by: Zilberstein, Noam
Published: (2024)

SeedLM: Compressing LLM Weights into Seeds of Pseudo-Random Generators
by: Shafipour, Rasoul, et al.
Published: (2024)

A New Dataset and Methodology for Malicious URL Classification
by: Schvartzman, Ilan, et al.
Published: (2024)

Thermodynamic Characteristics of a Fermi Gas with an Invariant Energy Scale and its Astrophysical Implications
by: Kar, Tiyasa, et al.
Published: (2026)

X-Token: Projection-Guided Cross-Tokenizer Knowledge Distillation
by: Sreenivas, Sharath Turuvekere, et al.
Published: (2026)

Localization and corruption : panacea or pandora's box? / Tugrul Gurgur, Anwar Shah
by: Gurgur, Tugrul
Published: (2005)

Stochastic Approximation with Two Time Scales: The General Case
by: Borkar, Vivek S
Published: (2024)