arXiv subjects

Po-han Li

Publications and source records attributed to Po-han Li.

18 recordsLinked to original sources

arXiv2026

ViewMind3D: Modular View-Aware Inference for Training-Free 3D-QA

Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled new possibilities for 3D question answering (3D-QA), a key capability for embodied AI and robotic perception. However, most existing methods rely on 3D-specific training or fine-tuning with costly annotations, limiting their scalability and real-world applicability. We present \textbf{ViewMind3D}, a fully training-free and modular framework for 3D spatial reasoning over multi-view observations of a scene without requiring complete 3D reconstruction. The framework decomposes the 3D-QA task into four interpretable components: (1) question-driven multi-view selection, (2) guided visual grounding with language-conditioned object cues, (3) spatial context encoding via a bird's-eye-view (BEV) viewpoint indicator, and (4) structured answer generation through role-based reasoning. This design enables structured, robust, and interpretable reasoning without requiring model tuning. Experimental results on ScanQA and SQA3D show that ViewMind3D achieves competitive performance compared to prior training-free and fine-tuned 3D-LLMs. In particular, our method improves performance on spatially grounded question types, such as ``What'' questions in SQA3D, while maintaining strong overall accuracy (50.8\%) and achieving 73.4 CIDEr on ScanQA. These results demonstrate that effective 3D reasoning can be achieved through modular orchestration of general-purpose LLMs and VLMs for robotic perception in real-world environments.

cs.CV↗

arXiv2026

VEGAS: Human-Aligned Video Caption Evaluation via Gaze

Vision-language models excel at video captioning, yet typically generate descriptions that fail to capture individual viewers' attention. We propose VEGAS (Video caption Evaluation via GAze Score), a training-free metric that leverages test-time gaze to sample personalized, attention-aligned text. It is a cross-modal, information-theoretic metric that quantifies how well a candidate caption matches a viewer's focus. To evaluate VEGAS, we curate a dataset of egocentric activities and instructional slides paired with synchronized gaze and reference annotations. We then select captions based on VEGAS via rejection sampling without model retraining. Experiments show that VEGAS-selected captions align significantly better with human focus and improve downstream caption-to-video retrieval, demonstrating the practical utility of incorporating viewer attention during inference.

cs.CV↗

arXiv2026

What We are Missing in Multimodal LLM Evaluation?

Multimodal large language models (MLLMs) can process diverse inputs, e.g., text, images, audio, and video, and generate textual responses. While their capabilities have advanced rapidly, evaluation of such models has not kept pace. Most existing evaluation benchmarks are limited to isolated tasks and reveal little about whether a model integrates information across modalities. We examine current means for evaluating MLLMs and review the existing benchmark taxonomy to identify gaps, including temporal-spatial coherence, physical world understanding, multimodal consistency, and selective attention. Addressing these gaps is essential for measuring real progress in multimodal intelligence and exposing capability boundaries.

cs.AI↗

arXiv2026

SSR: A Generic Framework for Text-Aided Map Compression for Localization

Mapping is crucial in robotics for localization and downstream decision-making. As robots are deployed in ever-broader settings, the maps they rely on continue to increase in size. However, storing these maps indefinitely (cold storage), transferring them across networks, or sending localization queries to cloud-hosted maps imposes prohibitive memory and bandwidth costs. We propose a text-enhanced compression framework that reduces both memory and bandwidth footprints while retaining high-fidelity localization. The key idea is to treat text as an alternative modality: one that can be losslessly compressed with large language models. We propose leveraging lightweight text descriptions combined with very small image feature vectors, which capture "complementary information" as a compact representation for the mapping task. Building on this, our novel technique, Similarity Space Replication (SSR), learns an adaptive image embedding in one shot that captures only the information "complementary" to the text descriptions. We validate our compression framework on multiple downstream localization tasks, including Visual Place Recognition as well as object-centric Monte Carlo localization in both indoor and outdoor settings. SSR achieves 2 times better compression than competing baselines on state-of-the-art datasets, including TokyoVal, Pittsburgh30k, Replica, and KITTI.

cs.CV↗

arXiv2026

ViSIL: Unified Evaluation of Information Loss in Multimodal Video Captioning

Multimodal video captioning condenses dense footage into a structured format of keyframes and natural language. By creating a cohesive multimodal summary, this approach anchors generative AI in rich semantic evidence and serves as a lightweight proxy for high-efficiency retrieval. However, traditional metrics like BLEU or ROUGE fail to quantify information coverage across disparate modalities, such as comparing a paragraph of text to a sequence of keyframes. To address this, we propose the Video Summary Information Loss (ViSIL) score, an information-theoretic framework that quantifies the video information not captured by a summary via vision-language model (VLM) inference. By measuring the information loss, ViSIL is a unified metric that enables direct comparison across multimodal summary formats despite their structural discrepancies. Our results demonstrate that ViSIL scores show a statistically significant correlation with both human and VLM performance on Video Question Answering (VQA) tasks. ViSIL also enables summary selection to optimize the trade-off between information loss and processing speed, establishing a Pareto-optimal frontier that outperforms text summaries by $7\%$ in VQA accuracy without increasing processing load.

cs.CV↗

arXiv2025

UNCAP: Uncertainty-Guided Neurosymbolic Planning Using Natural Language Communication for Cooperative Autonomous Vehicles

Safe large-scale coordination of multiple cooperative connected autonomous vehicles (CAVs) hinges on communication that is both efficient and interpretable. Existing approaches either rely on transmitting high-bandwidth raw sensor data streams or neglect perception and planning uncertainties inherent in shared data, resulting in systems that are neither scalable nor safe. To address these limitations, we propose Uncertainty-Guided Natural Language Cooperative Autonomous Planning (UNCAP), a vision-language model-based planning approach that enables CAVs to communicate via lightweight natural language messages while explicitly accounting for perception uncertainty in decision-making. UNCAP features a two-stage communication protocol: (i) an ego CAV first identifies the subset of vehicles most relevant for information exchange, and (ii) the selected CAVs then transmit messages that quantitatively express their perception uncertainty. By selectively fusing messages that maximize mutual information, this strategy allows the ego vehicle to integrate only the most relevant signals into its decision-making, improving both the scalability and reliability of cooperative planning. Experiments across diverse driving scenarios show a 63% reduction in communication bandwidth with a 31% increase in driving safety score, a 61% reduction in decision uncertainty, and a four-fold increase in collision distance margin during near-miss events. Project website: https://uncap-project.github.io/

cs.RO↗

arXiv2025

Fair Resource Allocation for Fleet Intelligence

Resource allocation is crucial for the performance optimization of cloud-assisted multi-agent intelligence. Traditional methods often overlook agents' diverse computational capabilities and complex operating environments, leading to inefficient and unfair resource distribution. To address this, we open-sourced Fair-Synergy, an algorithmic framework that utilizes the concave relationship between the agents' accuracy and the system resources to ensure fair resource allocation across fleet intelligence. We extend traditional allocation approaches to encompass a multidimensional machine learning utility landscape defined by model parameters, training data volume, and task complexity. We evaluate Fair-Synergy with advanced vision and language models such as BERT, VGG16, MobileNet, and ResNets on datasets including MNIST, CIFAR-10, CIFAR-100, BDD, and GLUE. We demonstrate that Fair-Synergy outperforms standard benchmarks by up to 25% in multi-agent inference and 11% in multi-agent learning settings. Also, we explore how the level of fairness affects the least advantaged, most advantaged, and average agents, providing insights for equitable fleet intelligence.

cs.LG↗

arXiv2025

VIBE: Annotation-Free Video-to-Text Information Bottleneck Evaluation for TL;DR

Many decision-making tasks, where both accuracy and efficiency matter, still require human supervision. For example, tasks like traffic officers reviewing hour-long dashcam footage or researchers screening conference videos can benefit from concise summaries that reduce cognitive load and save time. Yet current vision-language models (VLMs) often produce verbose, redundant outputs that hinder task performance. Existing video caption evaluation depends on costly human annotations and overlooks the summaries' utility in downstream tasks. We address these gaps with Video-to-text Information Bottleneck Evaluation (VIBE), an annotation-free method that scores VLM outputs using two metrics: grounding (how well the summary aligns with visual content) and utility (how informative it is for the task). VIBE selects from randomly sampled VLM outputs by ranking them according to the two scores to support effective human decision-making. Human studies on LearningPaper24, SUTD-TrafficQA, and LongVideoBench show that summaries selected by VIBE consistently improve performance-boosting task accuracy by up to 61.23% and reducing response time by 75.77% compared to naive VLM summaries or raw video.

cs.CV↗

arXiv2024

Dense Dynamics-Aware Reward Synthesis: Integrating Prior Experience with Demonstrations

Many continuous control problems can be formulated as sparse-reward reinforcement learning (RL) tasks. In principle, online RL methods can automatically explore the state space to solve each new task. However, discovering sequences of actions that lead to a non-zero reward becomes exponentially more difficult as the task horizon increases. Manually shaping rewards can accelerate learning for a fixed task, but it is an arduous process that must be repeated for each new environment. We introduce a systematic reward-shaping framework that distills the information contained in 1) a task-agnostic prior data set and 2) a small number of task-specific expert demonstrations, and then uses these priors to synthesize dense dynamics-aware rewards for the given task. This supervision substantially accelerates learning in our experiments, and we provide analysis demonstrating how the approach can effectively guide online learning agents to faraway goals.

cs.LG↗

arXiv2024

Any2Any: Incomplete Multimodal Retrieval with Conformal Prediction

Autonomous agents perceive and interpret their surroundings by integrating multimodal inputs, such as vision, audio, and LiDAR. These perceptual modalities support retrieval tasks, such as place recognition in robotics. However, current multimodal retrieval systems encounter difficulties when parts of the data are missing due to sensor failures or inaccessibility, such as silent videos or LiDAR scans lacking RGB information. We propose Any2Any-a novel retrieval framework that addresses scenarios where both query and reference instances have incomplete modalities. Unlike previous methods limited to the imputation of two modalities, Any2Any handles any number of modalities without training generative models. It calculates pairwise similarities with cross-modal encoders and employs a two-stage calibration process with conformal prediction to align the similarities. Any2Any enables effective retrieval across multimodal datasets, e.g., text-LiDAR and text-time series. It achieves a Recall@5 of 35% on the KITTI dataset, which is on par with baseline models with complete modalities.

cs.CV↗

arXiv2024

CSA: Data-efficient Mapping of Unimodal Features to Multimodal Features

Multimodal encoders like CLIP excel in tasks such as zero-shot image classification and cross-modal retrieval. However, they require excessive training data. We propose canonical similarity analysis (CSA), which uses two unimodal encoders to replicate multimodal encoders using limited data. CSA maps unimodal features into a multimodal space, using a new similarity score to retain only the multimodal information. CSA only involves the inference of unimodal encoders and a cubic-complexity matrix decomposition, eliminating the need for extensive GPU-based model training. Experiments show that CSA outperforms CLIP while requiring $50,000\times$ fewer multimodal data pairs to bridge the modalities given pre-trained unimodal encoders on ImageNet classification and misinformative news caption detection. CSA surpasses the state-of-the-art method to map unimodal features to multimodal features. We also demonstrate the ability of CSA with modalities beyond image and text, paving the way for future modality pairs with limited paired multimodal data but abundant unpaired unimodal data, such as lidar and text.

cs.LG↗

arXiv2024

Exploiting Distribution Constraints for Scalable and Efficient Image Retrieval

Image retrieval is crucial in robotics and computer vision, with downstream applications in robot place recognition and vision-based product recommendations. Modern retrieval systems face two key challenges: scalability and efficiency. State-of-the-art image retrieval systems train specific neural networks for each dataset, an approach that lacks scalability. Furthermore, since retrieval speed is directly proportional to embedding size, existing systems that use large embeddings lack efficiency. To tackle scalability, recent works propose using off-the-shelf foundation models. However, these models, though applicable across datasets, fall short in achieving performance comparable to that of dataset-specific models. Our key observation is that, while foundation models capture necessary subtleties for effective retrieval, the underlying distribution of their embedding space can negatively impact cosine similarity searches. We introduce Autoencoders with Strong Variance Constraints (AE-SVC), which, when used for projection, significantly improves the performance of foundation models. We provide an in-depth theoretical analysis of AE-SVC. Addressing efficiency, we introduce Single-shot Similarity Space Distillation ((SS)$_2$D), a novel approach to learn embeddings with adaptive sizes that offers a better trade-off between size and performance. We conducted extensive experiments on four retrieval datasets, including Stanford Online Products (SoP) and Pittsburgh30k, using four different off-the-shelf foundation models, including DinoV2 and CLIP. AE-SVC demonstrates up to a $16\%$ improvement in retrieval performance, while (SS)$_2$D shows a further $10\%$ improvement for smaller embedding sizes.

cs.IR↗

arXiv2024

PEERNet: An End-to-End Profiling Tool for Real-Time Networked Robotic Systems

Networked robotic systems balance compute, power, and latency constraints in applications such as self-driving vehicles, drone swarms, and teleoperated surgery. A core problem in this domain is deciding when to offload a computationally expensive task to the cloud, a remote server, at the cost of communication latency. Task offloading algorithms often rely on precise knowledge of system-specific performance metrics, such as sensor data rates, network bandwidth, and machine learning model latency. While these metrics can be modeled during system design, uncertainties in connection quality, server load, and hardware conditions introduce real-time performance variations, hindering overall performance. We introduce PEERNet, an end-to-end and real-time profiling tool for cloud robotics. PEERNet enables performance monitoring on heterogeneous hardware through targeted yet adaptive profiling of system components such as sensors, networks, deep-learning pipelines, and devices. We showcase PEERNet's capabilities through networked robotics tasks, such as image-based teleoperation of a Franka Emika Panda arm and querying vision language models using an Nvidia Jetson Orin. PEERNet reveals non-intuitive behavior in robotic systems, such as asymmetric network transmission and bimodal language model output. Our evaluation underscores the effectiveness and importance of benchmarking in networked robotics, demonstrating PEERNet's adaptability. Our code is open-source and available at github.com/UTAustin-SwarmLab/PEERNet.

cs.RO↗

arXiv2024

Online Foundation Model Selection in Robotics

Foundation models have recently expanded into robotics after excelling in computer vision and natural language processing. The models are accessible in two ways: open-source or paid, closed-source options. Users with access to both face a problem when deciding between effective yet costly closed-source models and free but less powerful open-source alternatives. We call it the model selection problem. Existing supervised-learning methods are impractical due to the high cost of collecting extensive training data from closed-source models. Hence, we focus on the online learning setting where algorithms learn while collecting data, eliminating the need for large pre-collected datasets. We thus formulate a user-centric online model selection problem and propose a novel solution that combines an open-source encoder to output context and an online learning algorithm that processes this context. The encoder distills vast data distributions into low-dimensional features, i.e., the context, without additional training. The online learning algorithm aims to maximize a composite reward that includes model performance, execution time, and costs based on the context extracted from the data. It results in an improved trade-off between selecting open-source and closed-source models compared to non-contextual methods, as validated by our theoretical analysis. Experiments across language-based robotic tasks such as Waymo Open Dataset, ALFRED, and Open X-Embodiment demonstrate real-world applications of the solution. The results show that the solution significantly improves the task success rate by up to 14%.

cs.RO↗

arXiv2023

Task-aware Distributed Source Coding under Dynamic Bandwidth

Efficient compression of correlated data is essential to minimize communication overload in multi-sensor networks. In such networks, each sensor independently compresses the data and transmits them to a central node due to limited communication bandwidth. A decoder at the central node decompresses and passes the data to a pre-trained machine learning-based task to generate the final output. Thus, it is important to compress the features that are relevant to the task. Additionally, the final performance depends heavily on the total available bandwidth. In practice, it is common to encounter varying availability in bandwidth, and higher bandwidth results in better performance of the task. We design a novel distributed compression framework composed of independent encoders and a joint decoder, which we call neural distributed principal component analysis (NDPCA). NDPCA flexibly compresses data from multiple sources to any available bandwidth with a single model, reducing computing and storage overhead. NDPCA achieves this by learning low-rank task representations and efficiently distributing bandwidth among sensors, thus providing a graceful trade-off between performance and bandwidth. Experiments show that NDPCA improves the success rate of multi-view robotic arm manipulation by 9% and the accuracy of object detection tasks on satellite imagery by 14% compared to an autoencoder with uniform bandwidth allocation.

cs.IT↗

arXiv2023

Differentially Private Timeseries Forecasts for Networked Control

We analyze a cost-minimization problem in which the controller relies on an imperfect timeseries forecast. Forecasting models generate imperfect forecasts because they use anonymization noise to protect input data privacy. However, this noise increases the control cost. We consider a scenario where the controller pays forecasting models incentives to reduce the noise and combines the forecasts into one. The controller then uses the forecast to make control decisions. Thus, forecasting models face a trade-off between accepting incentives and protecting privacy. We propose an approach to allocate economic incentives and minimize costs. We solve a biconvex optimization problem on linear quadratic regulators and compare our approach to a uniform incentive allocation scheme. The resulting solution reduces control costs by 2.5 and 2.7 times for the synthetic timeseries and the Uber demand forecast, respectively.

eess.SY↗

arXiv2023

Data Games: A Game-Theoretic Approach to Swarm Robotic Data Collection

Fleets of networked autonomous vehicles (AVs) collect terabytes of sensory data, which is often transmitted to central servers (the ''cloud'') for training machine learning (ML) models. Ideally, these fleets should upload all their data, especially from rare operating contexts, in order to train robust ML models. However, this is infeasible due to prohibitive network bandwidth and data labeling costs. Instead, we propose a cooperative data sampling strategy where geo-distributed AVs collaborate to collect a diverse ML training dataset in the cloud. Since the AVs have a shared objective but minimal information about each other's local data distribution and perception model, we can naturally cast cooperative data collection as an $N$-player mathematical game. We show that our cooperative sampling strategy uses minimal information to converge to a centralized oracle policy with complete information about all AVs. Moreover, we theoretically characterize the performance benefits of our game-theoretic strategy compared to greedy sampling. Finally, we experimentally demonstrate that our method outperforms standard benchmarks by up to $21.9\%$ on 4 perception datasets, including for autonomous driving in adverse weather conditions. Crucially, our experimental results on real-world datasets closely align with our theoretical guarantees.

cs.AI↗

arXiv2022

Adversarial Examples for Model-Based Control: A Sensitivity Analysis

We propose a method to attack controllers that rely on external timeseries forecasts as task parameters. An adversary can manipulate the costs, states, and actions of the controllers by forging the timeseries, in this case perturbing the real timeseries. Since the controllers often encode safety requirements or energy limits in their costs and constraints, we refer to such manipulation as an adversarial attack. We show that different attacks on model-based controllers can increase control costs, activate constraints, or even make the control optimization problem infeasible. We use the linear quadratic regulator and convex model predictive controllers as examples of how adversarial attacks succeed and demonstrate the impact of adversarial attacks on a battery storage control task for power grid operators. As a result, our method increases control cost by $8500\%$ and energy constraints by $13\%$ on real electricity demand timeseries.

eess.SY↗