Large language models have expanded beyond text to vision, audio, and other modalities, but understanding raw electromagnetic (EM) signals remains challenging. Wideband recordings can contain millions of samples, whereas existing waveform-based EM models are primarily evaluated on short in-phase and quadrature (IQ) segments, leaving long-recording understanding and detailed description insufficiently explored. We introduce LEMMO, a large EM model for instruction-driven recognition and description of long raw IQ recordings. LEMMO combines a linear-attention EM encoder with anti-aliased temporal resampling to process up to eight million complex samples. Its EM-text interaction module conditions learnable queries on instructions before retrieving signal evidence. Compact queries support recognition, while structured queries with auxiliary factual supervision organize scene-level properties, signal attributes, and cross-signal relationships for detailed description. We also construct and release a unified 1.38-TB corpus of real-world and synthetic EM signals, including LEMMO-ShortQA and LEMMO-LongDesc. Compared with MERLIN-LoRA, LEMMO improves mean EM-Bench perception accuracy by 15.77 %, mean strategy-generation ROUGE-L F1 by 4.38 points, and mean LEMMO-ShortQA accuracy by 14.31 %. It also improves categorical and numerical fact F1 on both LEMMO-LongDesc tasks while preserving its language backbone's MMLU performance. Our code and datasets are available at https://github.com/N1rv4n4/LEMMO and https://huggingface.co/datasets/DaftJun/RFLAS, respectively.
Figures & tables
Model
Raw IQ
Long IQ
Natural
Detailed Signal
UAV RF
Encoding
Sequence
Language QA
Description
Recognition
SpectrumFM
✓
✗
✗
✗
✗
EMind
✓
✗
✗
✗
✗
RF-GPT
✗
✗
✓
✓
✗
PReD
✗
✗
✓
✗
✗
MERLIN
✓
✗
✓
✗
✗
Table 1: Comparison of capabilities between existing EM models and LEMMO.
Figure 1: Overview of LEMMO. Raw IQ signals are encoded and temporally compressed for long recordings. Instruction-conditioned queries retrieve signal evidence for a LoRA-adapted LLM, supporting recognition with 16 queries and detailed description with 128 structured queries.
Dataset
Train
Validation
Test
IQ length
LEMMO-ShortQA
3,832 / 15,328
870 / 3,480
846 / 3,384
128–5M
LEMMO-LongDesc
16,103 / 165,328
856 / 8,813
652 / 802
0.4M–8M
EM-134K
95,679 / 120,829
10,587 / 13,278
–
512/1,024
EM-Bench
–
–
3,370 / 4,200
512/1,024
Table 2: Dataset summary. Each split reports the number of IQ recordings / instruction–response pairs. A dash indicates an unused split. IQ length is measured in complex samples, with M denoting one million.
Task
Metric
LEMMO
MERLIN -LoRA
GPT-6 Luna
DeepSeek V4.1-Flash
Gemini-3.5 Flash-Lite
Qwen3.8 Flash
Perception
CJR
Acc.
83.67
72.33
13.00
11.67
15.33
21.33
MOD
Acc.
36.00
27.00
0.00
1.33
9.33
10.33
Radar bandwidth
Acc.
58.33
27.00
0.00
0.00
0.33
1.33
Radar duty cycle
Acc.
54.33
30.67
1.67
0.67
14.67
11.00
Radar NP
Acc.
82.00
81.00
9.33
1.00
23.00
18.33
Table 3: Performance on the 14 EM-Bench tasks. Perception tasks report accuracy (Acc., %), and strategy-generation tasks report ROUGE-L F1 (R-L, scaled to 0–100). Averages are computed across tasks within each group. The highest score in each row is bold.
Model
Overall
STEM
Humanities
Social Sciences
Other
MERLIN
30.00
28.45
27.97
32.95
31.70
MERLIN-LoRA
57.55
53.60
48.33
70.26
62.92
Qwen3-4B-Instruct-2507
57.41
50.02
53.41
68.38
60.12
Qwen3.5-4B
64.34
58.77
59.70
71.86
69.58
LEMMO
64.98
60.04
59.49
72.64
70.71
Table 4: MMLU accuracy (%) for EM-adapted models and their original language backbones. MERLIN and MERLIN-LoRA use Qwen3-4B-Instruct-2507; LEMMO uses Qwen3.5-4B. The highest score in each column is bold.
Model
AMC
JPR
UAV
WTC
Average
GPT-6 Luna
0.32
4.00
0.00
0.08
1.10
DeepSeek-V4.1-Flash
0.81
3.12
0.00
1.95
1.47
Gemini-3.5-Flash-Lite
2.35
10.29
0.00
2.42
3.77
Qwen3.8-Flash
4.14
3.48
0.00
4.14
2.94
MERLIN-LoRA
60.01
61.47
41.88
59.61
55.74
LEMMO
69.72
69.24
73.75
67.50
70.05
Table 5: Recognition accuracy (%) on the LEMMO-ShortQA test set. Average denotes the unweighted mean across the four tasks. The highest score in each column is bold.
AMC
JPR
UAV
WTC
128–1,024 samples
128–16,368 samples
5,000,000 samples
4,096 samples
Table 6: Input IQ lengths for each LEMMO-ShortQA task, measured in complex samples.
Text Metrics
Fact Metrics
Model
R-1
R-2
R-L
BLEU-4
Token F1
Category F1
Value F1
Multi-signal scene overview task
GPT-6 Luna
22.79
6.26
11.49
3.86
23.51
0.00
0.00
DeepSeek-V4.1-Flash
21.88
5.95
10.36
2.64
23.01
0.00
0.50
Gemini-3.5-Flash-Lite
20.32
5.33
11.25
0.93
21.89
0.10
0.00
Qwen3.8-Flash
20.62
5.50
10.69
1.99
22.02
1.20
0.30
Table 7: Description performance on the LEMMO-LongDesc test set. All scores are scaled to 0–100. The highest score for each metric within each task is bold.
Input IQ length
4K
64K
256K
1M
4M
4K → 4M
Category F1
42.13
49.31
61.69
63.08
68.45
+26.32
Value F1
13.03
24.76
24.77
20.00
35.24
+22.21
Table 8: Effect of input IQ length on fact metrics. The same questions are evaluated at each length.
Appendix figures & tables13 assets
Supplementary material from the paper’s appendix.
Appendix
Dataset
Train signals / QA
Validation signals / QA
Test signals / QA
IQ length
Primary use
LEMMO-ShortQA
3,832 / 15,328
870 / 3,480
846 / 3,384
128-5M
Q16 Recoginition training and testing
LEMMO-LongDesc
16,103 / 165,328
856 / 8,813
652 / 802
0.4-8M
Q128 Description training and testing
EM-134K
95,679 / 120,829
10,587 / 13,278
–
512 or 1,024
EM-Bench training and validating
EM-Bench
–
–
3,370 / 4,200
512 or 1,024
EM-Bench testing
Appendix
Table 9: Signal and QA counts used in the reported experiments.
Task
Labels
Train signals / QA
Validation signals / QA
Test signals / QA
Modulation recognition
13
1,856 / 7,424
456 / 1,824
432 / 1,728
Jamming-pattern recognition
12
1,088 / 4,352
182 / 728
182 / 728
UAV recognition
10
120 / 480
40 / 160
40 / 160
Wireless-technology recognition
10
768 / 3,072
192 / 768
192 / 768
Total
45
3,832 / 15,328
870 / 3,480
846 / 3,384
Appendix
Table 10: LEMMO-ShortQA composition. Each physical signal has four question phrasings for the same task.
Large language models (LLMs) are increasingly used as code agents for scientific and engineering analysis, but their ability to analyze raw physical-layer measurements remains untested. We introduce EMRB (Electromagnetic Reasoning Benchmark), which evaluates whether LLMs can analyze raw I/Q data by writing and running code. EMRB contains 200 problems across five difficulty levels and 27 question types, from signal detection to OFDM design, generated from 11 signal types with verified ground truth. Unlike benchmarks built on preprocessed features or structured tables, EMRB provides only the raw capture; the quantities each question refers to must first be discovered through code. We evaluate 14 LLMs spanning proprietary, open-weight, and reasoning-oriented families. Scores range from 24.1% to 78.9%, with the mean dropping from 84.9% on basic measurement to 21.2% on system design. We also propose ReconPilot, a structured method that separates signal reconnaissance, targeted analysis, and self-verification. Across three backbones, ReconPilot raises the overall score by 3.8 to 17.6 points and improves 13 of 15 backbone-level combinations tested. All data and code are publicly released in blue{our GitHub repository}.
Mingxu Zhang, Ying Sun, Yuhan Li +4
The Hong Kong University of Science and Technology (Guangzhou) · The 63rd Research Institute, National University of Defense Technology · Nanjing University of Aeronautics and Astronautics
The paradigm of Multimodal Large Language Models (MLLMs) offers a promising blueprint for advancing the electromagnetic (EM) domain. However, prevailing approaches often deviate from the native MLLM paradigm, instead using task-specific or pipelined architectures that lead to fundamental limitations in model performance and generalization. Fully realizing the MLLM potential in EM domain requires overcoming three main challenges: (1) Data. The scarcity of high-quality datasets with paired EM signals and descriptive text annotations used for MLLMs pre-training; (2) Benchmark. The absence of comprehensive benchmarks to systematically evaluate and compare the performance of models on EM signal-to-text tasks; (3) Model. A critical fragility in low Signal-to-Noise Ratio (SNR) environments, where critical signal features can be obscured, leading to significant performance degradation. To address these challenges, we introduce a tripartite contribution to establish a foundation for MLLMs in the EM domain. First, to overcome data scarcity, we construct and release EM-100k, a large-scale dataset comprising over 100,000 EM signal-text pairs. Second, to enable rigorous and standardized evaluation, we propose EM-Bench, the most comprehensive benchmark featuring diverse downstream tasks spanning from perception to reasoning. Finally, to tackle the core modeling challenge, we present MERLIN, a novel training framework designed not only to align low-level signal representations with high-level semantic text, but also to explicitly enhance model robustness and performance in challenging low-SNR environments. Comprehensive experiments validate our method, showing that MERLIN is state-of-the-art in the EM-Bench and exhibits remarkable robustness in low-SNR settings.
Junyu Shen, Zhendong She, Chenghanyu Zhang +13
Tsinghua University · Tianjin University · Beijing University of Posts and Telecommunications +6
Electromagnetic spectrum monitoring increasingly requires flexible analysis beyond task-specific recognition and detection. Multimodal large language models offer a unified interface, but extending vision-language models (VLMs) to raw I/Q signals requires tokenization that balances fidelity against a strict budget. For signals, dense encoding causes token costs to grow with observation length, whereas fixed-resolution compression may discard short-duration or localized signal evidence. Thus, we propose BATok, a budget-adaptive signal tokenizer that adjusts token capacity to the input length while allocating that capacity according to the signal content. BATok constructs candidate representations from signal-derived features using lightweight multi-resolution branches, then combines a local energy prior with learnable queries to resample these representations into compact signal tokens. The number of tokens adapts to the input length while remaining strictly bounded. The resulting tokens are projected into the language embedding space of VLMs. We further introduce EMSpec-Instruct, a multimodal instruction dataset aligning raw I/Q signals, waterfall images, and language supervision for modulation recognition, structured detection, and language-conditioned signal grounding. Experiments show that BATok learns effective signal representations and achieves competitive performance across all tasks.
Lei Zhai, Zhihao Chang, Shuyuan Yang +1
School of Artificial Intelligence, Xidian University, Xi’an, China