M2BeamLLM combines four sensing modalities with a lightly fine-tuned GPT-2 backbone and reports 68.9% top-1 beam prediction accuracy on DeepSense 6G Scenario 32, beating the compared baselines by up to 13.9 percentage points.
LiDAR aided futur e beam prediction in real-world millimeter wave V2I communicatio ns,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
M2BeamLLM: Multimodal Sensing-empowered mmWave Beam Prediction with Large Language Models
M2BeamLLM combines four sensing modalities with a lightly fine-tuned GPT-2 backbone and reports 68.9% top-1 beam prediction accuracy on DeepSense 6G Scenario 32, beating the compared baselines by up to 13.9 percentage points.