Peng Jiang
3 papers · Latest:
Information Retrieval
Position-Aware Drafting for Inference Acceleration in LLM-Based Generative List-Wise Recommendation
PAD-Rec accelerates LLM-based generative recommendation by using position-aware speculative drafting, achieving up to 3.1x speedup.
2604.27747
RoboticsPushing Radar Odometry Beyond the Pavement: Current Capabilities and Challenges
This paper investigates off-road radar odometry, detailing challenges and proposing Radar-KISSICP and Radar-IMU baselines for improved estimation.
2604.24674
Information RetrievalCS3: Efficient Online Capability Synergy for Two-Tower Recommendation
CS3 is an efficient online framework that enhances two-tower recommender systems by improving representation, alignment, and cross-feature interactions.
2604.19269
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.