# Goodput Maximization for Large Language Model Edge Inference: A Two-Phase Maskable PPO Approach

arXiv: https://arxiv.org/abs/2608.25543 · HTML: https://mlsanity.com/paper/2608.25543/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 13 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 6 of 6)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [KIVI: A Tuning-Free Asymmetric 2bit Quantization for KV Cache](https://mlsanity.com/paper/2402.02750/) (2024)
- [DistServe: Disaggregating Prefill and Decoding for Goodput-optimized Large Language Model Serving](https://mlsanity.com/paper/2401.09670/) (2024)
- [Resource Allocation for Multiuser Edge Inference with Batching and Early Exiting (Extended Version)](https://mlsanity.com/paper/2204.05223/) (2022)
- [Generative AI as a Service in 6G Edge-Cloud: Generation Task Offloading by In-context Learning](https://mlsanity.com/paper/2408.02549/) (2024)
- [End-to-End Delay Minimization based on Joint Optimization of DNN Partitioning and Resource Allocation for Cooperative Edge Inference](https://mlsanity.com/paper/2310.12937/) (2023)

## Abstract (arXiv)

This paper presents a novel two-phase maskable proximal policy optimization (TP-MPPO) algorithm, which maximizes the system goodput counting request throughput with strict service level objective (SLO) compliance for large language model (LLM) inference services in wireless edge networks. In the first phase of TP-MPPO, we optimize the task offloading decisions by MPPO with action masking mechanism, effectively avoiding exploring invalid actions and reducing the action space. In the second phase, closed-form solutions are derived for uplink bandwidth allocation; a greedy algorithm is designed for downlink bandwidth allocation to provide immediate rewards for the MPPO in the next round. The two stages alternate till convergence. Simulation results demonstrate that TP-MPPO can improve the system reward by 33.3%--87.5% compared to its benchmarks and achieve the highest goodput.