Tubular trees, such as blood vessels and lung airways, are essential for material transport within the human body. Accurately detecting their centerlines with correct tree topology is critical for clinical tasks such as diagnosis, treatment planning, and surgical navigation. In these applications, maintaining high recall is crucial, as missing small branches can result in fatal mistakes caused by incomplete assessments or undetected abnormalities. We present RefTr, a 3D image-to-graph model for centerline generation of vascular trees via recurrent refinement of confluent trajectories. RefTr uses a Producer-Refiner architecture based on a Transformer decoder, where the Producer proposes a set of initial confluent trajectories that are recurrently refined by the Refiner to produce final trajectories, which forms the centerline graph. The confluent trajectory representation enables refinement of complete trajectories while explicitly enforcing a valid tree topology. The recurrent refinement scheme improves precision and reuses the same Refiner block across multiple steps, yielding a 2.4x reduction in decoder parameters compared to previous SOTA. We also introduce an efficient non-maximum suppression algorithm for spatial tree graphs to merge duplicate branches and boost precision. Across multiple public centerline datasets, RefTr achieves superior recall and comparable precision to previous SOTA, while offering faster inference and substantially fewer parameters, demonstrating its potential as a new state-of-the-art framework for vascular tree analysis in 3D medical imaging.

State-of-the-art extractive question-answering models achieve superhuman performances on the SQuAD benchmark. Yet, they are unreasonably heavy and need expensive GPU computing to answer questions in a reasonable time. Thus, they cannot be used in the open-domain question-answering paradigm for real-world queries on hundreds of thousands of documents. In this paper, we explore the possibility of transferring the natural language understanding of language models into dense vectors representing questions and answer candidates to make question-answering compatible with a simple nearest neighbor search task. This new model, which we call EfficientQA, takes advantage of the pair of sequences kind of input of BERT-based models to build meaningful, dense representations of candidate answers. These latter are extracted from the context in a question-agnostic fashion. Our model achieves state-of-the-art results in Phrase-Indexed Question Answering (PIQA), beating the previous state-of-art by 1.3 points in exact-match and 1.4 points in f1-score. These results show that dense vectors can embed rich semantic representations of sequences, although these were built from language models not originally trained for the use case. Thus, to build more resource-efficient NLP systems in the future, training language models better adapted to build dense representations of phrases is one of the possibilities.

Table extraction from document images is a challenging AI problem, and labelled data for many content domains is difficult to come by. Existing table extraction datasets often focus on scientific tables due to the vast amount of academic articles that are readily available, along with their source code. However, there are significant layout and typographical differences between tables found across scientific, financial, and other domains. Current datasets often lack the words, and their positions, contained within the tables, instead relying on unreliable OCR to extract these features for training modern machine learning models on natural language processing tasks. Therefore, there is a need for a more general method of obtaining labelled data. We present SynFinTabs, a large-scale, labelled dataset of synthetic financial tables. Our hope is that our method of generating these synthetic tables is transferable to other domains. To demonstrate the effectiveness of our dataset in training models to extract information from table images, we create FinTabQA, a layout large language model trained on an extractive question-answering task. We test our model using real-world financial tables and compare it to a state-of-the-art generative model and discuss the results. We make the dataset, model, and dataset generation code publicly available.

This paper presents a trajectory optimization and control approach for the guidance of an orbital four-arm robot in extravehicular activities. The robot operates near the target spacecraft, enabling its arm's end-effectors to reach the spacecraft's surface. Connections to the target spacecraft can be established by the arms through specific footholds (docking devices). The trajectory optimization allows the robot path planning by computing the docking positions on the target spacecraft surface, along with their timing, the arm trajectories, the six degrees of freedom body motion, and the necessary contact forces during docking. In addition, the paper introduces a controller designed to track the planned trajectories derived from the solution of the nonlinear programming problem. A weighted controller formulated as a convex optimization problem is proposed. The controller is defined as the optimization of an objective function that allows the system to perform a set of tasks simultaneously. Simulation results show the application of the trajectory optimization and control approaches to an on-orbit servicing scenario.

While large multimodal models excel in broad vision-language benchmarks, they often struggle with tasks requiring precise perception of low-level visual details, such as comparing line lengths or solving simple mazes. In particular, this failure mode persists in question-answering tasks about vector graphics -- images composed purely of 2D objects and shapes. To address this challenge, we propose the Visually Descriptive Language Model (VDLM), which performs text-based reasoning about vector graphics. VDLM leverages Scalable Vector Graphics (SVG) for a more precise visual description and first uses an off-the-shelf raster-to-SVG algorithm for encoding. Since existing language models cannot understand raw SVGs in a zero-shot setting, VDLM then bridges SVG with pretrained language models through a newly introduced intermediate symbolic representation, Primal Visual Description (PVD), comprising primitive attributes (e.g., shape, position, measurement) with their corresponding predicted values. PVD is task-agnostic and represents visual primitives that are universal across all vector graphics. It can be learned with procedurally generated (SVG, PVD) pairs and also enables the direct use of LLMs for generalization to complex reasoning tasks. By casting an image to a text-based representation, we can leverage the power of language models to learn alignment from SVG to visual primitives and generalize to unseen question-answering tasks. Empirical results show that VDLM achieves stronger zero-shot performance compared to state-of-the-art LMMs, such as GPT-4V, in various low-level multimodal perception and reasoning tasks on vector graphics. We additionally present extensive analyses on VDLM's performance, demonstrating that our framework offers better interpretability due to its disentangled perception and reasoning processes. Project page: https://mikewangwzhl.github.io/VDLM/

In this study, to perform the robotic stroke motions following the shape of the human back similar to the stroke motions by humans, in contrast to the conventional robotic stroke motion with a linear trajectory, we propose a trajectory generation method for a robotic stroke motion following the shape of the human back. We confirmed that the accuracy of the method's trajectory was close to that of the actual stroking motion by a human. Furthermore, we conducted a subjective experiment to evaluate the psychological effects of the proposed stroke motion in contrast to those of the conventional stroke motion with a linear trajectory. The experimental results showed that the actual stroke motion following the shape of the human back tended to evoke more pleasant and active feelings than the conventional stroke motion.

Sentiment analysis is a natural language processing task that aims to identify and extract the emotional aspects of a text. However, many existing sentiment analysis methods primarily classify the overall polarity of a text, overlooking the specific phrases that convey sentiment. In this paper, we applied an approach to sentiment analysis based on a question-answering framework. Our approach leverages the power of Bidirectional Autoregressive Transformer (BART), a pre-trained sequence-to-sequence model, to extract a phrase from a given text that amplifies a given sentiment polarity. We create a natural language question that identifies the specific emotion to extract and then guide BART to pay attention to the relevant emotional cues in the text. We use a classifier within BART to predict the start and end positions of the answer span within the text, which helps to identify the precise boundaries of the extracted emotion phrase. Our approach offers several advantages over most sentiment analysis studies, including capturing the complete context and meaning of the text and extracting precise token spans that highlight the intended sentiment. We achieved an end loss of 87% and Jaccard score of 0.61.

We propose a novel multi-section cable-driven soft robotic arm inspired by octopus tentacles along with a new modeling approach. Each section of the modular manipulator is made of a soft tubing backbone, a soft silicon arm body, and two rigid endcaps, which connect adjacent sections and decouple the actuation cables of different sections. The soft robotic arm is made with casting after the rigid endcaps are 3D-printed, achieving low-cost and convenient fabrication. To capture the nonlinear effect of cables pushing into the soft silicon arm body, which results from the absence of intermediate rigid cable guides for higher compliance, an analytical static model is developed to capture the relationship between the bending curvature and the cable lengths. The proposed model shows superior prediction performance in experiments over that of a baseline model, especially under large bending conditions. Based on the nonlinear static model, a kinematic model of a multi-section arm is further developed and used to derive a motion planning algorithm. Experiments show that the proposed soft arm has high flexibility and a large workspace, and the tracking errors under the algorithm based on the proposed modeling approach are up to 52$\%$ smaller than those with the algorithm derived from the baseline model. The presented modeling approach is expected to be applicable to a broad range of soft cable-driven actuators and manipulators.

BERT has shown a lot of sucess in a wide variety of NLP tasks. But it has a limitation dealing with long inputs due to its attention mechanism. Longformer, ETC and BigBird addressed this issue and effectively solved the quadratic dependency problem. However we find that these models are not sufficient, and propose LittleBird, a novel model based on BigBird with improved speed and memory footprint while maintaining accuracy. In particular, we devise a more flexible and efficient position representation method based on Attention with Linear Biases (ALiBi). We also show that replacing the method of global information represented in the BigBird with pack and unpack attention is more effective. The proposed model can work on long inputs even after being pre-trained on short inputs, and can be trained efficiently reusing existing pre-trained language model for short inputs. This is a significant benefit for low-resource languages where large amounts of long text data are difficult to obtain. As a result, our experiments show that LittleBird works very well in a variety of languages, achieving high performance in question answering tasks, particularly in KorQuAD2.0, Korean Question Answering Dataset for long paragraphs.

In this paper, we present our approach and empirical observations for Cause-Effect Signal Span Detection -- Subtask 2 of Shared task 3~\cite{tan-etal-2022-event} at CASE 2022. The shared task aims to extract the cause, effect, and signal spans from a given causal sentence. We model the task as a reading comprehension (RC) problem and apply a token-level RC-based span prediction paradigm to the task as the baseline. We explore different training objectives to fine-tune the model, as well as data augmentation (DA) tricks based on the language model (LM) for performance improvement. Additionally, we propose an efficient beam-search post-processing strategy to due with the drawbacks of span detection to obtain a further performance gain. Our approach achieves an average $F_1$ score of 54.15 and ranks \textbf{$1^{st}$} in the CASE competition. Our code is available at \url{https://github.com/Gzhang-umich/1CademyTeamOfCASE}.