CVPR Findings2026
BitTP: The Lightweight Trajectory Prediction Model with BitLLM for Edge-Devices
Mincheol Kang, HyunJin Lim, Bomin Kang, Daehee Park
Abstract
Trajectory prediction is a fundamental task for autonomous systems, requiring complex reasoning about multi-agent interactions and intents. Large language models (LLMs) have recently been adopted for this task, as they provide strong contextual reasoning and interpretable, language-based trajectory representations. However, these LLM-based predictors are extremely memoryand compute-intensive, making them difficult to deploy on resource-constrained edge devices such as on-board computers in autonomous robots. To bridge this gap, we propose BitTP, which converts an LLM-based trajectory predictor into a lightweight bitlinear architecture. We demonstrate that weight-only quantization to 1.58-bit (BitTP-Weight) is optimal. Crucially, activations must remain in full precision, as quantizing them leads to severe degradation and instability in spatio-temporal reasoning. Empirically, BitTP-Weight not only preserves but improves prediction quality over the full-precision (BF16) LLM baseline, reducing ADE by 14.29% and FDE by 20.97% on average, while simultaneously reducing memory usage and inference latency relative to other quantization methods. These results demonstrate that carefully designed quantization acts as an effective regularizer, enabling the practical deployment of sophisticated LLM-based reasoning on edge devices. Code is available at: https://github.com/ MintCat98/BitTP.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
- QLoRA: Efficient Finetuning of Quantized LLMs cites LoRA: Low-Rank Adaptation of Large Language Models
- AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent Forecasting cites ALBERT: A Lite BERT for Self-supervised Learning of Language Representations
Builds on18
- LoRA: Low-Rank Adaptation of Large Language ModelsEdward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu et al.ICLR 2022 · 18,833 citations
- ALBERT: A Lite BERT for Self-supervised Learning of Language RepresentationsZhenzhong Lan, Mingda Chen, Sebastian Goodman, Kevin Gimpel et al.ICLR 2020 · 7,418 citations
- QLoRA: Efficient Finetuning of Quantized LLMsTim Dettmers, Artidoro Pagnoni, Ari Holtzman, Luke ZettlemoyerNeurIPS 2023 · 5,863 citations
- Large Scale Interactive Motion Forecasting for Autonomous Driving : The Waymo Open Motion DatasetScott Ettinger, Shuyang Cheng, Benjamin Caine, Chenxi Liu et al.ICCV 2021 · 817 citations
- AgentFormer: Agent-Aware Transformers for Socio-Temporal Multi-Agent ForecastingYe Yuan, Xinshuo Weng, Yanglan Ou, Kris KitaniICCV 2021 · 658 citations
Related papers
- MobileQuant: Mobile-friendly Quantization for On-device Language ModelsFuwen Tan, Royson Lee, Lukasz Dudziak, Shell Xu Hu et al.EMNLP Findings 2024 · 13 citations
- Agile-Quant: Activation-Guided Quantization for Faster Inference of LLMs on the EdgeXuan Shen, Peiyan Dong, Lei Lu, Zhenglun Kong et al.AAAI 2024 · 59 citations
- BBAL: A Bidirectional Block Floating Point-Based Quantisation Accelerator for Large Language ModelsXiaomeng Han, Yuan Cheng, Jing Wang, Junyang Lu et al.DAC 2025 · 5 citations
- SPEED-Q: Staged Processing with Enhanced Distillation Towards Efficient Low-Bit On-Device VLM QuantizationTianyu Guo, Shanwei Zhao, Shiai Zhu, Chenguang MaAAAI 2026
- LBLLM: Lightweight Binarization of Large Language Models via Three-Stage DistillationSiqing Song, Chuang Wang, Yong Lang, Yi Yang et al.ACL 2026
