Institution profile

Imperial College London

Academic institutioneurope · gb
Official website
Research library2,327linked papers
Opportunities0open roles
Selected work

Representative Papers

A Survey of Deep Face Restoration: Denoise, Super-Resolution, Deblur, Artifact Removal

Nov 05, 2022arXiv.org

This paper presents a systematic survey of deep learning–based facial image restoration, focusing on denoising, super-resolution, deblurring, and artifact removal. Addressing challenges such as strong facial structural priors and complex degradation modeling, we propose the first holistic taxonomy of methods, a unified evaluation framework, and an open-source benchmark repository encompassing 20+ state-of-the-art approaches—including fully reproducible implementations. Leveraging datasets like CelebA and FFHQ, we conduct comprehensive cross-method evaluations using PSNR, SSIM, and LPIPS metrics, integrating CNN/Transformer architectures, perceptual and adversarial losses, and multi-scale feature fusion strategies. Our empirical analysis reveals performance boundaries and task-specific suitability across methods. Key contributions include: (1) the first structured, principle-driven classification system for facial restoration; (2) the first open, end-to-end benchmark platform supporting full method reproduction; (3) a rigorous, large-scale empirical study; and (4) concrete research directions concerning network design, evaluation paradigms, and dataset construction.

40 citations1 influentialRead paper

What is the Role of Small Models in the LLM Era: A Survey

Sep 10, 2024arXiv.org

Despite the dominance of large language models (LLMs), small models (SMs) remain indispensable in resource-constrained settings due to their low computational cost and deployment flexibility—yet their structural role has been systematically undervalued. Method: This paper introduces the first “collaboration–competition” two-dimensional analytical framework to characterize the dynamic interplay between SMs and LLMs, substantiated through systematic literature review, multi-case empirical comparison, and open-source implementation (GitHub repository). Contribution/Results: The study rigorously delineates SMs’ applicability boundaries across model compression, edge deployment, and human-AI collaboration. It delivers an actionable tripartite guideline covering model selection strategies, lightweight optimization techniques, and deployment paradigms—challenging the misconception that SMs are merely degraded substitutes for LLMs and affirming their foundational complementary role within the AGI ecosystem.

19 citationsRead paper

Bridging Language and Action: A Survey of Language-Conditioned Robot Manipulation

Dec 17, 2023

This work addresses the semantic gap between natural language instructions and robotic physical actions to enhance the naturalness and reliability of human-robot collaboration. We propose the first four-dimensional taxonomy for language-conditioned robotic manipulation—comprising reward shaping, policy learning, neurosymbolic AI, and foundation model–driven approaches—and systematically analyze their fundamental limitations in generalization and safety. Integrating large language models (LLMs), vision-language models (VLMs), neurosymbolic reasoning, and multimodal semantic parsing, we develop a unified analytical framework spanning semantic extraction, environmental assessment, and auxiliary task design. Our analysis rigorously characterizes the performance boundaries of each paradigm for the first time, establishing theoretical foundations and concrete technical pathways toward safe, generalizable, and interpretable language-driven robotic systems.

10 citationsRead paper

Impartial Games: A Challenge for Reinforcement Learning

May 25, 2022arXiv.org

This work identifies the fundamental cause of generalization failure in AlphaZero-style reinforcement learning on impartial games (e.g., Nim): neural networks relying on local observations cannot implicitly learn global, non-local abstract functions—such as parity—that determine game-theoretic outcomes, and state-outcome correlations vanish identically. We construct an AlphaZero variant integrating self-play training, Monte Carlo Tree Search, and residual CNNs, augmented with state-masking analysis and systematic generalization diagnostics. For the first time, we rigorously establish that the decoupling between local observability and global win-loss determination constitutes the core bottleneck for RL success in such domains, while also linking this failure to data skew and label noise. Experiments show convergence on small Nim instances, but training efficiency collapses catastrophically with scale; moreover, value networks fail to infer outcomes from partial states, exhibiting markedly lower robustness than in biased games like Chess or Go.

6 citations1 influentialRead paper

Analog Computing with Hybrid Couplers and Phase Shifters

Mar 14, 2026arXiv.org

This work proposes a novel approach to efficiently implement arbitrary linear transformations in the analog domain using low-cost microwave components. By constructing microwave networks composed solely of hybrid couplers and phase shifters, the method directly performs matrix-vector multiplication without digital computation, thereby significantly reducing latency. The study establishes, for the first time, necessary and sufficient conditions under which such architectures can realize any linear transformation and provides systematic hardware design procedures tailored to discrete Fourier transform (DFT), Hadamard, and Haar transforms at arbitrary power-of-two scales. A microstrip-based 4×4 DFT prototype was fabricated to validate the theoretical framework, with experimental results showing excellent agreement with predictions, thus demonstrating the architecture’s potential for high-speed analog signal processing.

6 citationsRead paper
Recent publications

Latest Papers

Memorisation bias in medical AI

Sep 15, 2026

研究揭示了医疗AI模型因记忆训练数据中的患者历史记录而产生的'记忆偏差'问题，影响未来诊断准确性，并提出需改变现有模型训练和部署协议以缓解该风险。

0 citationsRead paper