arXiv subjects

Chu Li

Publications and source records attributed to Chu Li.

At least 19 recordsLinked to original sources

arXiv2026

MaterialVellum: Steering Generative AI by Leveraging Expressive Material Affordances to Form Design Constraints

We present MaterialVellum, an interaction approach for steering generative AI through material-inspired manipulation of design constraints. In urban design, for instance, zoning bounds can be defined by the stretch of elastic bands, building density redistributed by molding clay, and functional relationships organized by the pull of magnets. MaterialVellum departs from current interfaces that rely on natural language prompts, and instead leverages users' existing knowledge of physical materials to represent and manipulate design constraints. We introduce a framework that maps material affordances---such as elasticity, malleability, or attraction---to spatial, relational, and proportional constraints that underpin design. These mappings are operationalized in user interfaces through familiar direct manipulation multi-touch gestures, including stretching, sculpting, and clustering, whose effects feed as control parameters into generative AI pipelines. We demonstrate this approach through technology probes, including an urban design, a flower arrangement design, and a graphical design scenario.A qualitative study (N=10) shows that material-based interactions improve accessibility of constraint expression, support exploratory workflows, and lower barriers for non-experts, while effectiveness depends on alignment with users' mental models.

cs.HC↗

arXiv2026

LandmarkLens: Predicting and Presenting Effective Landmarks for Mixed-Reality Urban Exploration

People with a poor sense of direction (SOD) struggle to build cognitive maps for effective spatial navigation, and existing navigation tools prioritize efficiency over spatial learning. To understand how navigation strategies differ by ability, we conducted a landmark attention study with 20 participants (ten good SOD, ten poor SOD) who navigated across four Tokyo neighborhoods in virtual reality (VR). We found systematic group differences in both gaze behavior and the types of landmarks they verbally identify as effective. Based on these findings, we built LandmarkLens, a mixed-reality (MR) navigation system that uses a vision-language model (VLM) to identify and highlight navigation-relevant landmarks. A follow-up study with eight poor-SOD participants showed improved performance in scene recognition, suggesting that guided landmark attention can support landmark-level spatial knowledge acquisition for people with poor SOD, a first step toward broader spatial learning.

cs.HC↗

arXiv2026

Douyin Multimodal Embedding Model Technical Report

Multimodal representation learning is a cornerstone of modern AI. By encoding multimodal queries and targets into vectors, it powers industrial search and recommendation and underpins modern agents. Real-world platforms with complex modalities and massive-scale content, such as Douyin, Xiaohongshu, and YouTube, demand both efficiency under billion-scale indexing and fine-grained discrimination for hard matching. Existing MLLM embedding models rarely satisfy both. Contrastive models are efficient but rely on pair-level supervision too coarse for fine-grained distinctions, while CoT-based models improve discrimination through explicit generation impractical to serve online. We present Douyin Multimodal Embedding (DME), a model trained in two stages to combine both strengths. Stage 1 performs large-scale contrastive pre-training that establishes a unified multimodal embedding space with broad modality and task coverage. Stage 2 supplements semantic sufficiency, the property that an embedding is grounded in retrieval-relevant evidence and preserves fine-grained counterpart-side semantics, via two mechanisms. Evidence-Grounded Typed Latent Reasoning organizes retrieval evidence through hidden-space latent reasoning, and Cross-Conditional Reconstruction enforces counterpart-side semantics through cross-directional autoregressive reconstruction. Both act only during training and add only marginal query-side overhead, so DME serves as efficiently as a standard contrastive encoder. On MMEB-v2, DME reaches state-of-the-art results at comparable scales for its 2B and 9B variants (74.8 and 78.4), with especially strong video and visual-document tasks. In production, DME delivers a 2.92% relative gain on Douyin's in-house offline evaluation set, is deployed across Douyin scenarios such as generative, image, and AI search, and yields a 0.1% Lifetime (LT) gain in online A/B testing on Douyin search.

cs.IR↗

arXiv2026

DynTrace: Tracking Dynamic Object Evidence for 4D Spatio-Temporal Reasoning in MLLMs

4D spatio-temporal reasoning, jointly modeling 3D spatial structure and temporal evolution, is essential for understanding dynamic worlds and enabling embodied interaction. While current Multimodal Large Language Models (MLLMs) show strong capabilities in static scene understanding and coarse-grained 4D tasks, they still have notable limitations in continuous dynamic scene perception, especially in tracking dynamic object evidence for coherent 4D spatio-temporal reasoning. This shortcoming stems mainly from relying on sparse frame-level observations, fragmenting continuous dynamic cues and leaving models unable to disentangle genuine object dynamics from camera-induced apparent motion. Inspired by humans tracking dynamic cues while compensating for viewpoint changes, we propose DynTrace, a training-free framework for 4D spatio-temporal reasoning with two complementary components. Dynamic Trajectory Visualization (DTV) reprojects world-coordinate trajectories onto the image plane, providing geometry-informed visual priors that disentangle genuine object dynamics from camera-induced apparent motion. Meanwhile, the Dynamic Trace Token (DT-Token), organized into a Dynamic Trace Graph (DTG), tracks object-level dynamic cues, trace evolution, and key moments, maintaining continuous dynamic object evidence for coherent 4D reasoning. Together, these two components equip MLLMs with continuously tracked dynamic object evidence, grounded in geometry-informed visual priors and structured spatio-temporal traces. DynTrace consistently improves open-source MLLMs, achieving state-of-the-art results on Dyn-Bench, VLM4D, and DSI-Bench, validating the importance of tracking dynamic object evidence for robust 4D spatio-temporal reasoning.

cs.CV↗

arXiv2026

Various electronic crystal phases in rhombohedral graphene multilayers

We systematically investigate the emergence of electron crystal phases in rhombohedral multilayer graphene using comprehensive self-consistent Hartree Fock calculations combined with \textit{ab initio} tight binding model. As the carrier density increases, we uncover an isospin cascade sequence of phase transitions that gives rise to a rich variety of ordered states, including electron crystal phases with non-zero Chern numbers. We further show the nearly degeneracy of these topological electron crystals hosting extended quantum anomalous Hall effect (EQAH) in the mean field regime and characterize pressure driven phase transitions. Finally, we discuss the thermodynamic signatures, particularly the behavior of the inverse compressibility, in light of recent experimental observations.

cond-mat.mes-hall↗

arXiv2026

Exploiting Near-Field Dynamics with Movable Antennas to Enhance Discrete Transmissive RIS

The design of low-complexity transceivers is crucial for the deployment of next-generation wireless systems. In this work, we combine two emerging concepts, movable antennas (MA) and transmissive reconfigurable intelligent surfaces (TRIS), which have recently attracted significant attention for enhancing wireless communication performance. In particular, we propose a compact base station (BS) architecture that integrates a single MA with a TRIS operating in their near-field region. We address the joint optimization of the MA location and the quantized TRIS phase configuration. Due to the non-convex coupling between spatial positioning and discrete phase constraints, an alternating optimization (AO) framework is developed, where the MA position is updated via gradient ascent (GA) and the TRIS phases are optimized through quantized phase alignment. Simulation results demonstrate that the proposed architecture significantly outperforms conventional BS designs equipped with fixed fully-active antenna arrays under the same channel model and transmit power constraint. Moreover, MA repositioning effectively mitigates the performance degradation caused by discrete TRIS phase quantization in near-field propagation environments. This reveals a favorable trade-off between hardware complexity and spatial signal processing, where the spatial adaptability of the MA can compensate for low-resolution TRIS phase control.

eess.SP↗

arXiv2026

GeoVisA11y: An AI-based Geovisualization Question-Answering System for Screen-Reader Users

Geovisualizations are powerful tools for communicating spatial information, but are inaccessible to screen-reader users. To address this limitation, we present GeoVisA11y, an LLM-based question-answering system that makes geovisualizations accessible through natural language interaction. The system supports map reading, analysis, interpretation and navigation by handling analytical, geospatial, visual and contextual queries. Through user studies with 12 screen-reader users and sighted participants, we demonstrate that GeoVisA11y effectively bridges accessibility gaps while revealing distinct interaction patterns between user groups. We contribute: (1) an open-source, accessible geovisualization system, (2) empirical findings on query and navigation differences, and (3) a dataset of geospatial queries to inform future research on accessible data visualization.

cs.HC↗

arXiv2026

U-MARVEL: Unveiling Key Factors for Universal Multimodal Retrieval via Embedding Learning with MLLMs

Universal multimodal retrieval (UMR), which aims to address complex retrieval tasks where both queries and candidates span diverse modalities, has been significantly advanced by the emergence of MLLMs. While state-of-the-art MLLM-based methods in the literature predominantly adopt contrastive learning principles, they often differ in their specific training recipes. Despite their success, the mechanisms underlying their retrieval capabilities remain largely unexplored, potentially resulting in suboptimal performance and limited generalization ability. To address these issues, we present a comprehensive study aimed at uncovering the key factors that drive effective embedding learning for UMR using MLLMs. We begin by implementing a general MLLM-based embedding learning pipeline, and systematically analyze the primary contributors to high-performing universal retrieval systems. Based on this, we explore various aspects of the details in embedding generation and training strategies, including progressive transition, hard negative mining and re-ranker distillation. Notably, our findings reveal that often-overlooked factors can have a substantial impact on model performance. Building on these discoveries, we introduce a unified framework termed U-MARVEL (Universal MultimodAl RetrieVal via Embedding Learning), which outperforms state-of-the-art competitors on the M-BEIR benchmark by a large margin in supervised settings, and also exhibits strong zero-shot performance on several tasks such as composed image retrieval and text-to-video retrieval. These results underscore the generalization potential of our framework across various embedding-based retrieval tasks. Code is available at https://github.com/chaxjli/U-MARVEL

cs.IR↗

arXiv2026

GenPairX: A Hardware-Algorithm Co-Designed Accelerator for Paired-End Read Mapping

Genome sequencing has become a central focus in computational biology. A genome study typically begins with sequencing, which produces millions to billions of short DNA fragments known as reads. Read mapping aligns these reads to a reference genome. Read mapping for short reads comes in two forms: single-end and paired-end, with the latter being more prevalent due to its higher accuracy and support for advanced analysis. Read mapping remains a major performance bottleneck in genome analysis due to expensive dynamic programming. Prior efforts have attempted to mitigate this cost by employing filters to identify and potentially discard computationally expensive matches and leveraging hardware accelerators to speed up the computations. While partially effective, these approaches have limitations. In particular, existing filters are often ineffective for paired-end reads, as they evaluate each read independently and exhibit relatively low filtering ratios. In this work, we propose GenPairX, a hardware-algorithm co-designed accelerator that efficiently minimizes the computational load of paired-end read mapping while enhancing the throughput of memory-intensive operations. GenPairX introduces: (1) a novel filtering algorithm that jointly considers both reads in a pair to improve filtering effectiveness, and a lightweight alignment algorithm to replace most of the computationally expensive dynamic programming operations, and (2) two specialized hardware mechanisms to support the proposed algorithms. Our evaluations show that GenPairX delivers substantial performance improvements over state-of-the-art solutions, achieving 1575x and 1.43x higher throughput per watt compared to leading CPU-based and accelerator-based read mappers, respectively, all without compromising accuracy.

cs.AR↗

arXiv2025

A Lightweight Framework for Integrated Sensing and Communications with RIS

Reconfigurable Intelligent Surfaces (RIS) have been recognized as a promising technology to enhance both communication and sensing performance in integrated sensing and communication (ISAC) systems for future 6G networks. However, existing RIS optimization methods for improving ISAC performance are mainly based on semidefinite relaxation (SDR) or iterative algorithms. The former suffers from high computational complexity and limited scalability, especially when the number of RIS elements becomes large, while the latter yields suboptimal solutions whose performance depends on initialization. In this work, we introduce a lightweight RIS phase design framework that provides a closed-form solution and explicitly accounts for the trade-off between communication and sensing, as well as proportional beam gain distribution toward multiple sensing targets. The key idea is to partition the RIS configuration into two parts: the first part is designed to maximize the communication performance, while the second introduces small perturbations to generate multiple beams for multi-target sensing. Simulation results validate the effectiveness of the proposed approach and demonstrate that it achieves performance comparable to SDR but with significantly lower computational complexity.

eess.SP↗

arXiv2025

Optimizing Movable Antenna Position and Transmissive RIS Phase for Efficient Base Station Design

Movable antennas (MA) and transmissive reconfigurable intelligent surfaces (TRIS) represent two innovative technologies that significantly enhance the flexibility of wireless communication systems. In this paper, we propose a novel and compact base station architecture that synergistically integrates a movable antenna with a transmissive RIS in the near field, enabling joint optimization of antenna positioning and TRIS phase adjustments. The proposed model compensates for phase quantization loss and significantly enhances signal strength, even with low-resolution (1-2 bit) phase shifters. Leveraging this framework, we systematically evaluate system performance as a function of TRIS size and antenna placement. Our results indicate that antenna mobility provides an additional degree of freedom to enhance the desired signal and achieve a higher SNR, particularly when combined with TRIS capabilities. These findings demonstrate that MA-TRIS integration offers a cost-effective and energy-efficient pathway toward compact 6G base stations, combining hardware simplicity with strong performance gains.

eess.SP↗

arXiv2025

"Where Can I Park?" Understanding Human Perspectives and Scalably Detecting Disability Parking from Aerial Imagery

Accessible parking is critical for people with disabilities (PwDs), allowing equitable access to destinations, independent mobility, and community participation. Despite mandates, there has been no large-scale investigation of the quality or allocation of disability parking in the US nor significant research on PwD perspectives and uses of disability parking. In this paper, we first present a semi-structured interview study with 11 PwDs to advance understanding of disability parking uses, concerns, and relevant technology tools. We find that PwDs often adapt to disability parking challenges according to their personal mobility needs and value reliable, real-time accessibility information. Informed by these findings, we then introduce a new deep learning pipeline, called AccessParkCV, and parking dataset for automatically detecting disability parking and inferring quality characteristics (e.g., width) from orthorectified aerial imagery. We achieve a micro-F1=0.89 and demonstrate how our pipeline can support new urban analytics and end-user tools. Together, we contribute new qualitative understandings of disability parking, a novel detection pipeline and open dataset, and design guidelines for future tools.

cs.HC↗

arXiv2025

Two-dimensional moiré phonon polaritons

Phonon polaritons (PhPs) are hybrid light-matter modes. We investigate them in two-dimensional (2D) materials with twisted moiré structures, revealing that the moiré potential creates a new class of `moiré PhPs'. These exhibit a fundamental spectral reconstruction into multiple branches and, crucially, electromagnetic wavefunctions that are nano-patterned by the superlattice. Through numerical simulations based on realistic lattice models, we confirm the existence of these intriguing modes. The inherent nanoscale structuring produces a robust, spatially varying near-field response, establishing moiré superlattices as a platform for engineering light-matter interactions.

cond-mat.mes-hall↗

arXiv2025

FlyMeThrough: Human-AI Collaborative 3D Indoor Mapping with Commodity Drones

Indoor mapping data is crucial for routing, navigation, and building management, yet such data are widely lacking due to the manual labor and expense of data collection, especially for larger indoor spaces. Leveraging recent advancements in commodity drones and photogrammetry, we introduce FlyMeThrough -- a drone-based indoor scanning system that efficiently produces 3D reconstructions of indoor spaces with human-AI collaborative annotations for key indoor points-of-interest (POI) such as entrances, restrooms, stairs, and elevators. We evaluated FlyMeThrough in 12 indoor spaces with varying sizes and functionality. To investigate use cases and solicit feedback from target stakeholders, we also conducted a qualitative user study with five building managers and five occupants. Our findings indicate that FlyMeThrough can efficiently and precisely create indoor 3D maps for strategic space planning, resource management, and navigation.

cs.HC↗

arXiv2025

Interactive Reasoning: Visualizing and Controlling Chain-of-Thought Reasoning in Large Language Models

The output quality of large language models (LLMs) can be improved via "reasoning": generating segments of chain-of-thought (CoT) content to further condition the model prior to producing user-facing output. While these chains contain valuable information, they are verbose and lack explicit organization, making them tedious to review. Moreover, they lack opportunities for user feedback, such as to remove unwanted considerations, add desired ones, or clarify unclear assumptions. We introduce Interactive Reasoning, an interaction design that visualizes chain-of-thought outputs as a hierarchy of topics and enables user review and modification. We implement interactive reasoning in Hippo, a prototype for AI-assisted decision making in the face of uncertain trade-offs. In a user study with 16 participants, we find that interactive reasoning in Hippo allows users to quickly identify and interrupt erroneous generations, efficiently steer the model towards customized responses, and better understand both model reasoning and model outputs. Our work contributes to a new paradigm that incorporates user oversight into LLM reasoning processes.

cs.HC↗

arXiv2025

Optimized Frequency-Diverse Movable Antenna Arrays for Directional Secrecy in Wireless Systems

Movable-antenna (MA) arrays are envisioned as a promising technique for enhancing secrecy performance in wireless communications by leveraging additional spatial degrees of freedom. However, when the eavesdropper is located in the same direction as the legitimate user, particularly in mmWave/THz bands where line-of-sight (LOS) propagation dominates, the secrecy performance of MA arrays becomes significantly limited, thus directionally insecure. To address this challenge, we employ a joint design that combines an MA array with a frequency-diverse array (FDA) at the transmitter to secure the transmission across both range and direction. Specifically, we derive closed-form expressions for the optimal antenna positions and frequency shifts, assuming small perturbations in both parameters from a linear frequency-diverse MA configuration. Furthermore, we compare the worst-case secrecy rate under this minor perturbation assumption with that obtained under a general constraint, where simulated annealing is employed to numerically determine the optimal parameters. Simulation results confirm that the proposed optimized frequency diverse MA approach significantly enhances secrecy performance in the presence of an eavesdropper aligned with the direction of the legitimate receiver.

eess.SP↗

arXiv2025

Accessibility for Whom? Perceptions of Sidewalk Barriers Across Disability Groups and Implications for Designing Personalized Maps

Despite diverse mobility needs worldwide, existing mapping tools fail to address the varied experiences of different mobility device users. This paper presents a large-scale online survey exploring how five mobility groups -- users of canes, walkers, mobility scooters, manual wheelchairs, and motorized wheelchairs -- perceive sidewalk barriers. Using 52 sidewalk barrier images, respondents evaluated their confidence in navigating each scenario. Our findings (N=190) reveal variations in barrier perceptions across groups, while also identifying shared concerns. To further demonstrate the value of this data, we showcase its use in two custom prototypes: a visual analytics tool and a personalized routing tool. Our survey findings and open dataset advance work in accessibility-focused maps, routing algorithms, and urban planning.

cs.HC↗

arXiv2024

Strong Inter-valley Electron-Phonon Coupling in Magic-Angle Twisted Bilayer Graphene

The unusual properties of superconductivity in magic-angle twisted bilayer graphene (MATBG) have sparked enormous research interest. However, despite the dedication of intensive experimental efforts and the proposal of several possible pairing mechanisms, the origin of its superconductivity remains elusive. Here, utilizing angle-resolved photoemission spectroscopy with micrometer spatial resolution, we have revealed flat band replicas in superconducting MATBG, where MATBG is unaligned with its hexagonal boron nitride (hBN) substrate11. These replicas exhibit uniform energy spacing, approximately 150 +- 15 meV apart, indicative of strong electron-boson coupling. Strikingly, these replicas are absent in non-superconducting twisted bilayer graphene (TBG) systems, either when MATBG is aligned to hBN or when TBG deviates from the magic angle. Calculations suggest that the formation of these flat band replicas in superconducting MATBG are attributed to the strong coupling between flat band electrons and an optical phonon mode at the graphene K point, facilitated by inter-valley scattering. These findings, although do not necessarily put electron phonon coupling as the main driving force for the superconductivity in MATBG, unravel the unique electronic structure inherent in superconducting MATBG, thereby providing crucial information for understanding the unusual electronic landscape from which the superconductivity is derived.

cond-mat.mes-hall↗