arXiv subjects

Lingfeng Huang

Publications and source records attributed to Lingfeng Huang.

3 recordsLinked to original sources

arXiv2026

Through Their Eyes: Fixation-aligned Tuning for Personalized User Emulation

Large language model (LLM) agents are increasingly deployed as scalable user simulators for recommender system evaluation. Yet existing simulators perceive recommendations through text or structured metadata rather than the visual interfaces real users browse-a critical gap, since attention over recommendation layouts is both visually driven and highly personalized. We investigate whether aligning a vision-language model's (VLM's) visual attention with user-specific gaze patterns can improve simulation fidelity. Analysis of a real-world eye-tracking dataset collected in a carousel-based recommendation setting reveals that users exhibit stable individual gaze patterns strongly predictive of click behavior. Building on this finding, we propose Fixation-Aligned Tuning for user Emulation (FixATE). Our approach first probes the VLM's internal visual attention via interpretability operators to obtain a slot-level relevance distribution comparable with human fixation, and then learns personalized soft prompts to steer the model's attention toward each user's characteristic fixation pattern. Experiments across three interpretability-based probing operators and two architecturally distinct VLM backbones demonstrate consistent improvements in both attention alignment and click prediction accuracy. These results suggest that making the model "see like the user" is a viable path toward simulators that more faithfully reproduce how users perceive and act in recommendation interfaces.

cs.MM↗

arXiv2025

RT-DEMT: A hybrid real-time acupoint detection model combining mamba and transformer

Traditional Chinese acupuncture methods often face controversy in clinical practice due to their high subjectivity. Additionally, current intelligent-assisted acupuncture systems have two major limitations: slow acupoint localization speed and low accuracy. To address these limitations, a new method leverages the excellent inference efficiency of the state-space model Mamba, while retaining the advantages of the attention mechanism in the traditional DETR architecture, to achieve efficient global information integration and provide high-quality feature information for acupoint localization tasks. Furthermore, by employing the concept of residual likelihood estimation, it eliminates the need for complex upsampling processes, thereby accelerating the acupoint localization task. Our method achieved state-of-the-art (SOTA) accuracy on a private dataset of acupoints on the human back, with an average Euclidean distance pixel error (EPE) of 7.792 and an average time consumption of 10.05 milliseconds per localization task. Compared to the second-best algorithm, our method improved both accuracy and speed by approximately 14\%. This significant advancement not only enhances the efficacy of acupuncture treatment but also demonstrates the commercial potential of automated acupuncture robot systems. Access to our method is available at https://github.com/Sohyu1/RT-DEMT

cs.CV↗

arXiv2024

Evaluating the potential of thermoplastic polymers for cryogenic sealing applications: strain rate and temperature effects

Cryogenic fuels, such as liquid hydrogen and liquid natural gas, emerge as versatile and sustainable energy carriers that are revolutionising various industries including aerospace, automotive, marine, and power generation. Thermoplastic polymers can be a suitable alternative to metal seals in cryogenic fuel systems. However, there is limited study about the behaviours of thermoplastics at cryogenic temperatures, especially at liquid hydrogen temperature of 20 Kelvin (K). This paper measured the tensile properties and coefficient of thermal expansion of three popular thermoplastics: PTFE, PEEK and UHMWPE at room temperature (RT), 77 K and 20 K and at four strain rates. Further microscopic analysis was also conducted to understand the failure mechanisms occurring when combining reduced temperature with varying strain rate. The tensile strength of each polymer increased from RT to 77 K and decreased from 77 K to 20 K. Elastic modulus tended to increase, and the strain recorded at failure decreased when reducing temperature from RT to 20 K. From microscopic observation of PEEK and UHMWPE, a reduction in temperature from 77 K to 20 K resulted in a larger instantaneous fracture, with multi-faceted fracture surfaces containing many small mirror like and opaque or misty sub-regions within the fracture zone. For PTFE, the surface morphology exhibited an insensitivity to the increase in strain rate at cryogenic temperatures, and the microscopy showed how the size of dimples found within the fracture interface became smaller when temperature was reduced from 77 K to 20 K. Finally, PEEK was found to contract much less than PTFE and UHMWPE at 20 K, in agreement to it having the highest glass transition temperature of the three polymers, which is normally a good indicator when attempting to identify polymers that will tend to exhibit smaller contraction at cryogenic temperatures.

cond-mat.mtrl-sci↗