Chinese Academy of Sciences

Selected work

Representative Papers

A Comprehensive Survey on Vector Database: Storage and Retrieval Technique, Challenge

Oct 18, 2023arXiv.org

Managing and retrieving high-dimensional vector data poses significant challenges, particularly as traditional databases fail to meet performance requirements and the need for tight integration with large language models (LLMs) intensifies. Method: This paper systematically surveys four major approximate nearest neighbor search (ANNS) paradigms—hashing, tree-based indexing, graph-based methods (e.g., HNSW), and quantization (PQ/SQ)—and integrates hybrid optimization strategies. Contribution/Results: It introduces, for the first time, a “Four-Dimensional Methodology” framework tailored for industrial deployment of vector databases, analyzing trade-offs among accuracy, latency, memory footprint, and scalability. The work constructs a structured knowledge graph covering 200+ ANNS algorithms and proposes a novel paradigm for deep synergy between vector databases and LLMs. Collectively, these contributions provide both theoretical foundations and practical guidelines for system selection, architectural design, and development of AI-native database systems.

62 citations3 influentialRead paper

Dynamic and Adaptive Feature Generation with LLM

Jun 04, 2024arXiv.org

Existing feature engineering approaches suffer from three fundamental limitations: poor interpretability, weak generalizability, and inflexible strategies—hindering practical deployment across diverse scenarios. To address these challenges, this paper proposes the first large language model (LLM)-driven dynamic adaptive feature generation paradigm. Our method integrates task-aware prompting with semantic modeling of the feature space, enabling real-time, interpretable, and controllable feature generation tailored to both data characteristics and task requirements. It ensures cross-modal and cross-task generality while maintaining full transparency in the feature generation process. Extensive experiments on multiple structured and unstructured data tasks demonstrate that features generated by our approach improve feature quality by 23.6% and boost downstream model performance by an average of 11.4%, significantly outperforming conventional automated feature engineering methods.

21 citations1 influentialRead paper

Learning Critically: Selective Self-Distillation in Federated Learning on Non-IID Data

Dec 01, 2024IEEE Transactions on Big Data

To address the challenges of poor generalization, slow convergence, and local model divergence from the global optimum in federated learning caused by non-IID data, this paper proposes a selective self-distillation framework. The method introduces a dual-level credibility assessment—operating at both class- and sample-level—to dynamically generate fine-grained self-distillation weights, enabling adaptive integration of global knowledge into local training. Crucially, it requires no auxiliary teacher model or additional communication overhead, while providing theoretical guarantees on convergence. Extensive experiments on three standard non-IID benchmark datasets demonstrate that the proposed approach significantly improves model generalization and robustness, outperforming existing state-of-the-art methods with fewer communication rounds.

20 citations3 influentialRead paper

Enhancing efficiency and propulsion in bio-mimetic robotic fish through end-to-end deep reinforcement learning

Mar 01, 2024The Physics of Fluids

Bionic robotic fish suffer from low propulsion efficiency and high energy consumption. Method: This study proposes an end-to-end deep reinforcement learning (DRL) control framework, introducing— for the first time in underwater bionic robotics—extended pressure sensing combined with temporal Transformer modeling, integrated with a policy transfer mechanism to enhance training stability and environmental adaptability. Training achieves autonomous, stable, and rapid convergence within CFD simulations (Re = 6000). Contribution/Results: The DRL policy improves propulsion efficiency by 37% and reduces specific energy consumption per unit thrust by 29% over conventional pre-programmed gaits. Flow-field analysis reveals that efficiency stems from embodied regulation of body deformation and vortex–body interactions. The core contribution is a novel bio-inspired locomotion control paradigm unifying perception, spatiotemporal modeling, and decision-making.

9 citationsRead paper

Continuous Speculative Decoding for Autoregressive Image Generation

Nov 18, 2024arXiv.org

Continuous autoregressive visual generation models suffer from high inference latency, while existing speculative decoding methods are restricted to discrete token spaces and lack theoretical foundations or practical techniques for continuous-valued outputs. Method: This work pioneers the extension of speculative decoding to continuous visual generation. We propose a diffusion-prior-based continuous acceptance criterion, design a denoising trajectory alignment mechanism and token pre-filling strategy to mitigate distribution mismatch, and establish a continuous accept-reject sampling framework with analytically derived upper bounds on approximation error. Contribution/Results: Our approach achieves a 2.33× inference speedup on standard diffusion-based autoregressive models while provably preserving the exact output distribution of the original model. The implementation is publicly available.

8 citationsRead paper

Recent publications