arXiv subjects

Tingting Hu

Publications and source records attributed to Tingting Hu.

5 recordsLinked to original sources

arXiv2026

A Birkhoff-Orthogonal Dehghan--Rooin Type Constant in Banach Spaces

For a real Banach space $X$. The Dehghan--Rooin magnitude compares the angular distance $α[x,y]=\|x/\|x\|-y/\|y\|\|$ together with the skew angular distance $β[x,y]=\|x/\|y\|-y/\|x\|\|$. Motivated by the recent use of Birkhoff orthogonality in the Massera--Schaffer inequality, we introduce and study the Birkhoff-restricted Dehghan--Rooin constant \[ \operatorname{DR}_{B}(X)=\sup\left\{\frac{α[x,y]}{β[x,y]}:x,y\in X\setminus\{0\},\ x\perp_{B} y,\ β[x,y]\ne0\right\}. \] The restriction $x\perp_{B} y$ turns the global comparison of $α$ and $β$ into a directional invariant which detects the interaction between supporting functionals and the radial normalization map. We obtain a scale-profile formula, sharp elementary bounds, stability under subspaces and ultrapowers, a variational characterization of the extremal case $\operatorname{DR}_{B}(X)=1$, estimates in terms of the classical Dehghan--Rooin constant, and consequences for uniformly convex and uniformly smooth spaces. In smooth two-dimensional spaces we derive an explicit formula through the normalized duality mapping, and in Radon planes we reduce the constant to a one-dimensional optimization on an arc of the unit circle. Several model computations are included, including Hilbert spaces, polyhedral planes, and two-dimensional $p$-norm spaces. The final section gives normal-structure type consequences through a companion Birkhoff angular modulus and lists several problems suggested by the new constant.

math.FA↗

arXiv2026

Guess Where You Go: Generative Next Point-of-Interest Recommendation in Amap

Generative retrieval enables recommender systems to retrieve items by generating compact item identifiers, but scaling it to industrial scenarios remains challenging due to redundant or colliding token assignments and insufficient integration of heterogeneous item signals. These challenges are particularly critical for next Point-of-Interest (POI) recommendation, where models must represent structured spatial entities, capture sequential mobility patterns, and produce predictions consistent with real user behavior. We propose Gwhere, an end-to-end industrial framework that integrates semantic identifier (SID) generation with LLM-based generative next POI recommendation. Gwhere first learns discriminative POI SIDs through a contrastive residual-quantization tokenizer that aligns textual, visual, spatial, and collaborative signals. Based on these SIDs, Gwhere adapts LLMs to mobility scenarios via continued pretraining on enriched spatio-temporal corpora, supervised fine-tuning, and Exposure-Aware Kahneman-Tversky Optimization (EAKTO), a reinforcement learning objective for behavioral preference alignment. Experiments on public datasets and Amap's large-scale industrial dataset demonstrate the effectiveness of Gwhere. The system has been deployed in Amap's homepage service under high-concurrency and low-latency constraints. Long-term online A/B tests show improvements of 5.83% in P-CTR and 6.20% in U-CTR over the production baseline. The implementation is publicly available at https://github.com/alibaba/SimCIT.

cs.IR↗

arXiv2025

Cognitive-Aligned Spatio-Temporal Large Language Models For Next Point-of-Interest Prediction

The next point-of-interest (POI) recommendation task aims to predict the users' immediate next destinations based on their preferences and historical check-ins, holding significant value in location-based services. Recently, large language models (LLMs) have shown great potential in recommender systems, which treat the next POI prediction in a generative manner. However, these LLMs, pretrained primarily on vast corpora of unstructured text, lack the native understanding of structured geographical entities and sequential mobility patterns required for next POI prediction tasks. Moreover, in industrial-scale POI prediction applications, incorporating world knowledge and alignment of human cognition, such as seasons, weather conditions, holidays, and users' profiles (such as habits, occupation, and preferences), can enhance the user experience while improving recommendation performance. To address these issues, we propose CoAST (Cognitive-Aligned Spatial-Temporal LLMs), a framework employing natural language as an interface, allowing for the incorporation of world knowledge, spatio-temporal trajectory patterns, profiles, and situational information. Specifically, CoAST mainly comprises of 2 stages: (1) Recommendation Knowledge Acquisition through continued pretraining on the enriched spatial-temporal trajectory data of the desensitized users; (2) Cognitive Alignment to align cognitive judgments with human preferences using enriched training data through Supervised Fine-Tuning (SFT) and a subsequent Reinforcement Learning (RL) phase. Extensive offline experiments on various real-world datasets and online experiments deployed in "Guess Where You Go" of AMAP App homepage demonstrate the effectiveness of CoAST.

cs.AI↗

arXiv2025

On the equivalent Zbǎganu constant associated with isosceles orthogonality in Banach spaces

This paper systematically investigates a new geometric constant associated with isosceles orthogonality in Banach spaces. By establishing the connection between this new constant and a classical function, sharp upper and lower bounds for the constant are derived. Specifically, the space is exactly a Hilbert space when the new constant reaches its lower bound; in finite-dimensional spaces, if the new constant attains its upper bound, it implies that the space does not possess uniform non-squareness.

math.FA↗

arXiv2024

Image registration based automated lesion correspondence pipeline for longitudinal CT data

Patients diagnosed with metastatic breast cancer (mBC) typically undergo several radiographic assessments during their treatment. mBC often involves multiple metastatic lesions in different organs, it is imperative to accurately track and assess these lesions to gain a comprehensive understanding of the disease's response to treatment. Computerized analysis methods that rely on lesion-level tracking have often used manual matching of corresponding lesions, a time-consuming process that is prone to errors. This paper introduces an automated lesion correspondence algorithm designed to precisely track both targets' lesions and non-targets' lesions in longitudinal data. Here we demonstrate the applicability of our algorithm on the anonymized data from two Phase III trials. The dataset contains imaging data of patients for different follow-up timepoints and the radiologist annotations for the patients enrolled in the trials. Target and non-target lesions are annotated by either one or two groups of radiologists. To facilitate accurate tracking, we have developed a registration-assisted lesion correspondence algorithm. The algorithm employs a sequential two-step pipeline: (a) Firstly, an adaptive Hungarian algorithm is used to establish correspondence among lesions within a single volumetric image series which have been annotated by multiple radiologists at a specific timepoint. (b) Secondly, after establishing correspondence and assigning unique names to the lesions, three-dimensional rigid registration is applied to various image series at the same timepoint. Registration is followed by ongoing lesion correspondence based on the adaptive Hungarian algorithm and updating lesion names for accurate tracking. Validation of our automated lesion correspondence algorithm is performed through triaxial plots based on axial, sagittal, and coronal views, confirming its efficacy in matching lesions.

eess.IV↗