arXiv subjects

Patrick Liu

Publications and source records attributed to Patrick Liu.

9 recordsLinked to original sources

arXiv2026

Spatial Reasoning is Not a Free Lunch: A Controlled Study on LLaVA

Vision-language models (VLMs) have advanced rapidly, yet they still struggle with basic spatial reasoning. Despite strong performance on general benchmarks, modern VLMs remain brittle at understanding 2D spatial relationships such as relative position, layout, and counting. We argue that this failure is not merely a data problem, but is closely tied to dominant design choices in current VLM pipelines: reliance on CLIP-style image encoders and the flattening of images into 1D token sequences with 1D positional encoding. We present a controlled diagnostic study within the LLaVA framework to isolate how these choices affect spatial grounding. We evaluate frontier models and LLaVA variants on a suite of spatial benchmarks, comparing CLIP-based encoders against alternatives trained with denser or generative objectives, as well as variants augmented with 2D positional encoding. Our results show consistent spatial performance gaps across models, and indicate that encoder objectives and positional structure shape spatial behavior, but do not fully resolve it.

cs.CV↗

arXiv2026

The Spatial Blindspot of Vision-Language Models

Vision-language models (VLMs) have advanced rapidly, but their ability to capture spatial relationships remains a blindspot. Current VLMs are typically built with contrastive language-image pretraining (CLIP) style image encoders. The training recipe often flattens images into 1D patch sequences, discarding the 2D structure necessary for spatial reasoning. We argue that this lack of spatial awareness is a missing dimension in VLM design and a bottleneck for applications requiring spatial grounding, such as robotics and embodied AI. To address this, we investigate (i) image encoders trained with alternative objectives and (ii) 2D positional encodings. Our experiments show that these architectural choices can lead to improved spatial reasoning on several benchmarks.

cs.CV↗

arXiv2025

Revisiting the charge-density-wave superlattice of 1$T$-TiSe$_2$

A number of intriguing phenomena, including exciton condensation, orbital ordering, and emergence of chirality, have been proposed to accompany charge-density-wave (CDW) formation in the layered transition metal dichalcogenide 1$T$-TiSe$_2$. Explaining these effects relies on knowledge of the atomic displacement pattern underlying the CDW, yet structural proposals based on spatially-averaging bulk crystal diffraction and surface-dependent scanning tunneling microscopy have remained inconsistent. Here, we revisit the CDW superlattice structure with selected-area electron diffraction, a bulk-sensitive probe capable of capturing sub-micrometer spatial variations while maintaining high momentum resolution. We resolved two distinct, spatially separated CDW phases characterized by different interlayer ordering. In both phases, previously reported atomic displacement patterns fail to account for the observed extinction rules. Instead, our analysis reveals a new superlattice structure, which features a large number of nearly degenerate CDW domains. These findings not only provide a new basis for understanding the gyrotropic electronic order and metastability in 1$T$-TiSe$_2$, they also underscore the importance of bulk-sensitive mesoscopic techniques in investigating materials that host unconventional superlattices.

cond-mat.str-el↗

arXiv2024

Geophysical Observations of the 24 September 2023 OSIRIS-REx Sample Return Capsule Re-Entry

Sample Return Capsules (SRCs) entering Earth's atmosphere at hypervelocity from interplanetary space are a valuable resource for studying meteor phenomena. The 24 September 2023 arrival of the OSIRIS-REx (Origins, Spectral Interpretation, Resource Identification, and Security-Regolith Explorer) SRC provided an unprecedented chance for geophysical observations of a well-characterized source with known parameters, including timing and trajectory. A collaborative effort involving researchers from 16 institutions executed a carefully planned geophysical observational campaign at strategically chosen locations, deploying over 400 ground-based sensors encompassing infrasound, seismic, distributed acoustic sensing (DAS), and GPS technologies. Additionally, balloons equipped with infrasound sensors were launched to capture signals at higher altitudes. This campaign (the largest of its kind so far) yielded a wealth of invaluable data anticipated to fuel scientific inquiry for years to come. The success of the observational campaign is evidenced by the near-universal detection of signals across instruments, both proximal and distal. This paper presents a comprehensive overview of the collective scientific effort, field deployment, and preliminary findings. The early findings have the potential to inform future space missions and terrestrial campaigns, contributing to our understanding of meteoroid interactions with planetary atmospheres. Furthermore, the dataset collected during this campaign will improve entry and propagation models as well as augment the study of atmospheric dynamics and shock phenomena generated by meteoroids and similar sources.

astro-ph.EP↗

arXiv2023

DiffusionSat: A Generative Foundation Model for Satellite Imagery

Diffusion models have achieved state-of-the-art results on many modalities including images, speech, and video. However, existing models are not tailored to support remote sensing data, which is widely used in important applications including environmental monitoring and crop-yield prediction. Satellite images are significantly different from natural images -- they can be multi-spectral, irregularly sampled across time -- and existing diffusion models trained on images from the Web do not support them. Furthermore, remote sensing data is inherently spatio-temporal, requiring conditional generation tasks not supported by traditional methods based on captions or images. In this paper, we present DiffusionSat, to date the largest generative foundation model trained on a collection of publicly available large, high-resolution remote sensing datasets. As text-based captions are sparsely available for satellite images, we incorporate the associated metadata such as geolocation as conditioning information. Our method produces realistic samples and can be used to solve multiple generative tasks including temporal generation, superresolution given multi-spectral inputs and in-painting. Our method outperforms previous state-of-the-art methods for satellite image generation and is the first large-scale generative foundation model for satellite imagery. The project website can be found here: https://samar-khanna.github.io/DiffusionSat/

cs.CV↗

arXiv2022

Enhancing Self-Consistency and Performance of Pre-Trained Language Models through Natural Language Inference

While large pre-trained language models are powerful, their predictions often lack logical consistency across test inputs. For example, a state-of-the-art Macaw question-answering (QA) model answers 'Yes' to 'Is a sparrow a bird?' and 'Does a bird have feet?' but answers 'No' to 'Does a sparrow have feet?'. To address this failure mode, we propose a framework, Consistency Correction through Relation Detection, or ConCoRD, for boosting the consistency and accuracy of pre-trained NLP models using pre-trained natural language inference (NLI) models without fine-tuning or re-training. Given a batch of test inputs, ConCoRD samples several candidate outputs for each input and instantiates a factor graph that accounts for both the model's belief about the likelihood of each answer choice in isolation and the NLI model's beliefs about pair-wise answer choice compatibility. We show that a weighted MaxSAT solver can efficiently compute high-quality answer choices under this factor graph, improving over the raw model's predictions. Our experiments demonstrate that ConCoRD consistently boosts accuracy and consistency of off-the-shelf closed-book QA and VQA models using off-the-shelf NLI models, notably increasing accuracy of LXMERT on ConVQA by 5% absolute. See https://ericmitchell.ai/emnlp-2022-concord/ for code and data.

cs.CL↗

arXiv2022

SatMAE: Pre-training Transformers for Temporal and Multi-Spectral Satellite Imagery

Unsupervised pre-training methods for large vision models have shown to enhance performance on downstream supervised tasks. Developing similar techniques for satellite imagery presents significant opportunities as unlabelled data is plentiful and the inherent temporal and multi-spectral structure provides avenues to further improve existing pre-training strategies. In this paper, we present SatMAE, a pre-training framework for temporal or multi-spectral satellite imagery based on Masked Autoencoder (MAE). To leverage temporal information, we include a temporal embedding along with independently masking image patches across time. In addition, we demonstrate that encoding multi-spectral data as groups of bands with distinct spectral positional encodings is beneficial. Our approach yields strong improvements over previous state-of-the-art techniques, both in terms of supervised learning performance on benchmark datasets (up to $\uparrow$ 7%), and transfer learning performance on downstream remote sensing tasks, including land cover classification (up to $\uparrow$ 14%) and semantic segmentation. Code and data are available on the project website: https://sustainlab-group.github.io/SatMAE/

cs.CV↗

arXiv2021

SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning

Progress toward the United Nations Sustainable Development Goals (SDGs) has been hindered by a lack of data on key environmental and socioeconomic indicators, which historically have come from ground surveys with sparse temporal and spatial coverage. Recent advances in machine learning have made it possible to utilize abundant, frequently-updated, and globally available data, such as from satellites or social media, to provide insights into progress toward SDGs. Despite promising early results, approaches to using such data for SDG measurement thus far have largely evaluated on different datasets or used inconsistent evaluation metrics, making it hard to understand whether performance is improving and where additional research would be most fruitful. Furthermore, processing satellite and ground survey data requires domain knowledge that many in the machine learning community lack. In this paper, we introduce SustainBench, a collection of 15 benchmark tasks across 7 SDGs, including tasks related to economic development, agriculture, health, education, water and sanitation, climate action, and life on land. Datasets for 11 of the 15 tasks are released publicly for the first time. Our goals for SustainBench are to (1) lower the barriers to entry for the machine learning community to contribute to measuring and achieving the SDGs; (2) provide standard benchmarks for evaluating machine learning models on tasks across a variety of SDGs; and (3) encourage the development of novel machine learning methods where improved model performance facilitates progress towards the SDGs.

cs.LG↗

arXiv2017

On the relative role of different age groups during epidemics associated with the respiratory syncytial virus

Background: While RSV circulation results in high burden of hospitalization, particularly among infants, young children and the elderly, little is known about the role of different age groups in propagating annual RSV epidemics in the community. Methods: During a communicable disease outbreak, some subpopulations may play a disproportionate role during the outbreak's ascent due to increased susceptibility and/or contact rates. Such subpopulations can be identified by considering the proportion that cases in a subpopulation represent among all cases in the population occurring before (Bp) and after the epidemic peak (Ap) to calculate the subpopulation's relative risk, RR=Bp/Ap. We estimated RR for several age groups using data on RSV hospitalizations in the US between 2001-2012 from the Healthcare Cost and Utilization Project (HCUP). Results: Children aged 3-4y and 5-6y each had the highest RR estimate for 5/11 seasons in the data, with RSV hospitalization rates in infants being generally higher during seasons when children aged 5-6y had the highest RR estimates. Children aged 2y had the highest RR estimate during one season. RR estimates in infants and individuals aged 11y and older were mostly lower than in children aged 1-10y. Conclusions: The RR estimates suggest that preschool and young school-age children have the leading relative roles during RSV epidemics. We hope that those results will aid in the design of RSV vaccination policies.

q-bio.PE↗