SatOps-Decision: Low-Latency Discriminative Decisions for Onboard Satellite Communication Operations

Lifan Mei; Yuxuan Li; Jiaxian Tu; Shenghe Xu

cs.LGOct 28, 2026

SatOps-Decision: Low-Latency Discriminative Decisions for Onboard Satellite Communication Operations

Authors: Lifan Mei, Yuxuan Li, Jiaxian Tu, Shenghe Xu

Organizations: Academy of AIAT, Xi’an Jiaotong-Liverpool University · Amazon, Inc.

Abstract

We propose SatOps-Decision, a method for low-latency decisions in satellite communication operations. To the best of our knowledge, it is the first application of Jev-style general-purpose discriminative decision models to satellite operations, and the accompanying benchmark is the first dedicated to this approach. The learned scorer, a locally adapted Laya encoder with its native decision head, selects among supplied operational answers in a single forward pass. The benchmark combines authored operational contracts, controlled state changes and independently checked labels in an auditable construction. Evaluation prioritizes complete-request latency and raw decision accuracy; all main accuracy comparisons use the same 1,024 held-out questions, separate from training and development. SatOps-Decision's local scorer has a median latency of 90.3 ms per decision, versus 493.9 ms for Qwen3.5-0.8B generation and 388.4 ms for OneJev1.6-0.8B: $5.47\times$ and $4.30\times$ speedups based on median request times in the measured FP32/MPS implementations. Its mean raw accuracy is 85.3%, versus 42.5% for frozen Laya, 40.7% for Qwen generation and 58.6% for OneJev. As cloud accuracy references, original Jev 1.14/1.13 reach 87.4/81.9%.

Figures & tables

Fig. 1: Satellite communication setting and four operational judgments. Evidence from $t_{m}$ is assessed at $t$ , with an answer at $t^{\prime}=t+\ell$ . Switching must fit within contact $[t_{s},t_{e})$ , reserve $D+\delta$ for the switch and margin, and meet the freshness and energy conditions in ( 4 ). Times are in seconds; energy uses synthetic percentage points. TDRS artwork: NASA/Goddard Space Flight Center [ 11 ] . Time axis is not to scale.

Fig. 2: Shared decision scorer and separate operational checks. Development data select the adapted weights; raw predictions are scored before the veto-only checker and kept separate from checked decisions.

Fig. 3: Laya neural architecture. All tokens traverse the encoder and decision head before extracting

k=|\mathcal{A}|

option vectors. Shared scoring and masked softmax yield candidate probabilities.

Training configuration	Updated parameters (M)	Raw accuracy (%)
Full adaptation (ours)	421.03	84.47
Head only	26.25	45.70
No order augmentation	421.03	84.67

TABLE IV: Adaptation controls on the standard test (seed 7).

LLM Decision-Making

LLM decision-making studies how language models choose, act, seek information, and allocate resources under goals, uncertainty, constraints, and social pressures.

See all 246 papers

StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams
Sep 29, 2026Jhen-Ke Lin, Chung Chun WangLLM Evaluation LLM Inference
National Yang Ming Chiao Tung University
System One Models for Wireless Decision-Making:Applications and Performance Evaluation
Oct 3, 2026Masoud Rahimi, S. M. Matin Alemohammad, Hamid Behroozi +1Wireless Resource Allocation
Department of Electrical Engineering, Sharif University of Technology, Tehran 1458889694, Iran
Jeva: System One Models from First Principles
Oct 13, 2026Qi Li, Xinchao WangLLM Decision-Making Efficient Language Model Inference
National University of Singapore

Sep 29, 2026J/K move · Enter open · S save

StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams

Jhen-Ke Lin, Chung Chun Wang

National Yang Ming Chiao Tung University

LLM Evaluation LLM Inference LLM Decision-Making Streaming Inference

Open paper AskPDF arXiv

Language models increasingly make real-time decisions in applications that apply the latest answer until a newer one arrives. A late answer can prolong an outdated decision, such as a call recorder still running while a customer reads out card details, an error offline accuracy misses. We make three contributions. First, we release StreamDecisionBench (SDB), a dataset of eight streaming scenarios in four application families, with executable reference decisions derived from public rules. Second, we propose an evaluation protocol and a metric, in-force accuracy: the share of time the applied decision is correct across update intervals of 0.5-8 s. It reflects accuracy and latency jointly, attributing each error to judgment, latency or both. Third, we evaluate fifteen single-model settings, and this attribution separates speed-limited from judgment-limited models: slower, more accurate models lose 42-47% of the time to outdated answers, a fast model 35% to wrong ones. We therefore test hybrids in which a slow model corrects a fast one; with the right pairing and configuration, a hybrid outperforms every single model. However, even the best evaluated system keeps a correct decision in force only about two-thirds of the time, leaving a substantial gap for real-time use.

Abstract

LLM Decision-Making

Explore similar work

StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams

System One Models for Wireless Decision-Making:Applications and Performance Evaluation

Jeva: System One Models from First Principles