Jinmeng Rao
YOU?
Author Swipe
View article: Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing
Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing Open
Retrieval Augmented Generation (RAG) has shown strong capability in enhancing language models' knowledge and reducing AI generative hallucinations, driving its widespread use. However, complex tasks requiring multi-round retrieval remain c…
View article: CRISP: Clustering Multi-Vector Representations for Denoising and Pruning
CRISP: Clustering Multi-Vector Representations for Denoising and Pruning Open
Multi-vector models, such as ColBERT, are a significant advancement in neural information retrieval (IR), delivering state-of-the-art performance by representing queries and documents by multiple contextualized token-level embeddings. Howe…
View article: Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing
Knowing You Don't Know: Learning When to Continue Search in Multi-round RAG through Self-Practicing Open
Retrieval Augmented Generation (RAG) has shown strong capability in enhancing language models' knowledge and reducing AI generative hallucinations, driving its widespread use. However, complex tasks requiring multi-round retrieval remain c…
View article: Towards the next generation of Geospatial Artificial Intelligence
Towards the next generation of Geospatial Artificial Intelligence Open
Geospatial Artificial Intelligence (GeoAI), as the integration of geospatial studies and AI, has become one of the fastest-developing research directions in spatial data science and geography. This rapid change in the field calls for a dee…
View article: MoLA: MoE LoRA with Layer-wise Expert Allocation
MoLA: MoE LoRA with Layer-wise Expert Allocation Open
View article: SRL: Towards a General-Purpose Framework for Spatial Representation Learning
SRL: Towards a General-Purpose Framework for Spatial Representation Learning Open
View article: IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues
IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues Open
View article: IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues
IM-RAG: Multi-Round Retrieval-Augmented Generation Through Learning Inner Monologues Open
Although the Retrieval-Augmented Generation (RAG) paradigms can use external knowledge to enhance and ground the outputs of Large Language Models (LLMs) to mitigate generative hallucinations and static knowledge base problems, they still s…
View article: Best Practices and Lessons Learned on Synthetic Data
Best Practices and Lessons Learned on Synthetic Data Open
The success of AI models relies on the availability of large, diverse, and high-quality datasets, which can be challenging to obtain due to data scarcity, privacy concerns, and high costs. Synthetic data has emerged as a promising solution…
View article: Tackling Vision Language Tasks through Learning Inner Monologues
Tackling Vision Language Tasks through Learning Inner Monologues Open
Visual language tasks such as Visual Question Answering (VQA) or Visual Entailment (VE) require AI models to comprehend and reason with both visual and textual content. Driven by the power of Large Language Models (LLMs), two prominent met…
View article: Higher Layers Need More LoRA Experts
Higher Layers Need More LoRA Experts Open
Parameter-efficient tuning (PEFT) techniques like low-rank adaptation (LoRA) offer training efficiency on Large Language Models, but their impact on model performance remains limited. Recent efforts integrate LoRA and Mixture-of-Experts (M…
View article: Performance of Linear Mixed Models in Estimating Structural Rates of Glaucoma Progression Using Varied Random Effect Distributions
Performance of Linear Mixed Models in Estimating Structural Rates of Glaucoma Progression Using Varied Random Effect Distributions Open
View article: Here Is Not There: Measuring Entailment-Based Trajectory Similarity for Location-Privacy Protection and Beyond
Here Is Not There: Measuring Entailment-Based Trajectory Similarity for Location-Privacy Protection and Beyond Open
While the paths humans take play out in social as well as physical space, measures to describe and compare their trajectories are carried out in abstract, typically Euclidean, space. When these measures are applied to trajectories of actua…
View article: Building Privacy-Preserving and Secure Geospatial Artificial Intelligence Foundation Models (Vision Paper)
Building Privacy-Preserving and Secure Geospatial Artificial Intelligence Foundation Models (Vision Paper) Open
In recent years we have seen substantial advances in foundation models for artificial intelligence, including language, vision, and multimodal models. Recent studies have highlighted the potential of using foundation models in geospatial a…
View article: FLEE-GNN
FLEE-GNN Open
Understanding and measuring the resilience of food supply networks is a\nglobal imperative to tackle increasing food insecurity. However, the complexity\nof these networks, with their multidimensional interactions and decisions,\npresents …
View article: CATS: Conditional Adversarial Trajectory Synthesis for privacy-preserving trajectory data publication using deep learning approaches
CATS: Conditional Adversarial Trajectory Synthesis for privacy-preserving trajectory data publication using deep learning approaches Open
The prevalence of ubiquitous location-aware devices and mobile Internet enables us to collect massive individual-level trajectory dataset from users. Such trajectory big data bring new opportunities to human mobility research but also rais…
View article: SSIF: Learning Continuous Image Representation for Spatial-Spectral Super-Resolution
SSIF: Learning Continuous Image Representation for Spatial-Spectral Super-Resolution Open
Existing digital sensors capture images at fixed spatial and spectral resolutions (e.g., RGB, multispectral, and hyperspectral images), and each combination requires bespoke machine learning models. Neural Implicit Functions partially over…
View article: CATS: Conditional Adversarial Trajectory Synthesis for Privacy-Preserving Trajectory Data Publication Using Deep Learning Approaches
CATS: Conditional Adversarial Trajectory Synthesis for Privacy-Preserving Trajectory Data Publication Using Deep Learning Approaches Open
The prevalence of ubiquitous location-aware devices and mobile Internet enables us to collect massive individual-level trajectory dataset from users. Such trajectory big data bring new opportunities to human mobility research but also rais…
View article: Evaluation and Enhancement of Semantic Grounding in Large Vision-Language Models
Evaluation and Enhancement of Semantic Grounding in Large Vision-Language Models Open
Large Vision-Language Models (LVLMs) offer remarkable benefits for a variety of vision-language tasks. However, a challenge hindering their application in real-world scenarios, particularly regarding safety, robustness, and reliability, is…
View article: Tackling Vision Language Tasks Through Learning Inner Monologues
Tackling Vision Language Tasks Through Learning Inner Monologues Open
Visual language tasks require AI models to comprehend and reason with both visual and textual content. Driven by the power of Large Language Models (LLMs), two prominent methods have emerged: (1) the hybrid integration between LLMs and Vis…
View article: LOWA: Localize Objects in the Wild with Attributes
LOWA: Localize Objects in the Wild with Attributes Open
We present LOWA, a novel method for localizing objects with attributes effectively in the wild. It aims to address the insufficiency of current open-vocabulary object detectors, which are limited by the lack of instance-level attribute cla…
View article: Improved Prediction of Perimetric Loss in Glaucomatous Eyes Using Latent Class Mixed Modeling
Improved Prediction of Perimetric Loss in Glaucomatous Eyes Using Latent Class Mixed Modeling Open
View article: Predictors of outcomes in patients with obesity following mitral valve surgery
Predictors of outcomes in patients with obesity following mitral valve surgery Open
The complexity of presentation, comorbidities in older and female patients, and morbid obesity are independently associated with an increased risk of mortality in patients undergoing open mitral valve replacement or repair. Morbid obesity …
View article: Disparities in Survival Due to Social Determinants of Health and Access to Treatment in US Patients With Operable Malignant Pleural Mesothelioma
Disparities in Survival Due to Social Determinants of Health and Access to Treatment in US Patients With Operable Malignant Pleural Mesothelioma Open
Importance Outcomes of localized malignant pleural mesothelioma (MPM) remain poor despite multimodality therapy. It is unclear what role disparities have in the overall survival (OS) of patients with operable MPM. Objective To examine surv…
View article: Choosing <scp>GIS</scp> graduate programs from afar: Chinese students' perspectives
Choosing <span>GIS</span> graduate programs from afar: Chinese students' perspectives Open
With the increasing demands for geospatial analytics in industry and academia, the need for Geographic Information Systems/Science (GIS) education is on the rise. A growing number of departments in geography have launched or expanded their…
View article: Measuring network resilience via geospatial knowledge graph
Measuring network resilience via geospatial knowledge graph Open
Quantifying the resilience in the food system is important for food security\nissues. In this work, we present a geospatial knowledge graph (GeoKG)-based\nmethod for measuring the resilience of a multi-commodity flow network.\nSpecifically…
View article: STICC: a multivariate spatial clustering method for repeated geographic pattern discovery with consideration of spatial contiguity
STICC: a multivariate spatial clustering method for repeated geographic pattern discovery with consideration of spatial contiguity Open
Spatial clustering has been widely used for spatial data mining and knowledge discovery. An ideal multivariate spatial clustering should consider both spatial contiguity and aspatial attributes. Existing spatial clustering approaches may f…
View article: Rates of Glaucoma Progression Derived from Linear Mixed Models Using Varied Random Effect Distributions
Rates of Glaucoma Progression Derived from Linear Mixed Models Using Varied Random Effect Distributions Open
Use of the LG distribution in models estimating rates of change among glaucoma patients may improve their accuracy in rapidly identifying progressors at high risk for vision loss.
View article: A Multi-perspective Narrative-Based Geovisualization Dashboard for the 2020 US Presidential Election
A Multi-perspective Narrative-Based Geovisualization Dashboard for the 2020 US Presidential Election Open
View article: Rates of Glaucoma Progression Derived from Linear Mixed Models Using Varied Random Effect Distributions
Rates of Glaucoma Progression Derived from Linear Mixed Models Using Varied Random Effect Distributions Open
Purpose To compare the ability of linear mixed models with different random effect distributions to estimate rates of visual field loss in glaucoma patients. Methods Eyes with ≥5 reliable standard automated perimetry (SAP) tests were ident…