VideoLLM Benchmarks and Evaluation: A Survey

The rapid development of Large Language Models (LLMs) has catalyzed significant advancements in video understanding technologies. This survey provides a comprehensive analysis of benchmarks and evaluation methodologies specifically designed or used for Video Large Language Models (VideoLLMs). We examine the current landscape of video understanding benchmarks, discussing their characteristics, evaluation protocols, and limitations. The paper analyzes various evaluation methodologies, including closed-set, open-set, and specialized evaluations for temporal and spatiotemporal understanding tasks. We highlight the performance trends of state-of-the-art VideoLLMs across these benchmarks and identify key challenges in current evaluation frameworks. Additionally, we propose future research directions to enhance benchmark design, evaluation metrics, and protocols, including the need for more diverse, multimodal, and interpretability-focused benchmarks. This survey aims to equip researchers with a structured understanding of how to effectively evaluate VideoLLMs and identify promising avenues for advancing the field of video understanding with large language models.

Design, Contact Modeling, and Collision-inclusive Planning of a Dual-stiffness Aerial RoboT (DART)

Collision-resilient quadrotors have gained significant attention given their potential for operating in cluttered environments and leveraging impacts to perform agile maneuvers. However, existing designs are typically single-mode: either safeguarded by propeller guards that prevent deformation or deformable but lacking rigidity, which is crucial for stable flight in open environments. This paper introduces DART, a Dual-stiffness Aerial RoboT, that adapts its post-collision response by either engaging a locking mechanism for a rigid mode or disengaging it for a flexible mode, respectively. Comprehensive characterization tests highlight the significant difference in post collision responses between its rigid and flexible modes, with the rigid mode offering seven times higher stiffness compared to the flexible mode. To understand and harness the collision dynamics, we propose a novel collision response prediction model based on the linear complementarity system theory. We demonstrate the accuracy of predicting collision forces for both the rigid and flexible modes of DART. Experimental results confirm the accuracy of the model and underscore its potential to advance collision-inclusive trajectory planning in aerial robotics.

Towards Making Flowchart Images Machine Interpretable

Computer programming textbooks and software documentations often contain flowcharts to illustrate the flow of an algorithm or procedure. Modern OCR engines often tag these flowcharts as graphics and ignore them in further processing. In this paper, we work towards making flowchart images machine-interpretable by converting them to executable Python codes. To this end, inspired by the recent success in natural language to code generation literature, we present a novel transformer-based framework, namely FloCo-T5. Our model is well-suited for this task,as it can effectively learn semantics, structure, and patterns of programming languages, which it leverages to generate syntactically correct code. We also used a task-specific pre-training objective to pre-train FloCo-T5 using a large number of logic-preserving augmented code samples. Further, to perform a rigorous study of this problem, we introduce theFloCo dataset that contains 11,884 flowchart images and their corresponding Python codes. Our experiments show promising results, and FloCo-T5 clearly outperforms related competitive baselines on code generation metrics. We make our dataset and implementation publicly available.

Improving Medical Multi-modal Contrastive Learning with Expert Annotations

We introduce eCLIP, an enhanced version of the CLIP model that integrates expert annotations in the form of radiologist eye-gaze heatmaps. It tackles key challenges in contrastive multi-modal medical imaging analysis, notably data scarcity and the "modality gap" -- a significant disparity between image and text embeddings that diminishes the quality of representations and hampers cross-modal interoperability. eCLIP integrates a heatmap processor and leverages mixup augmentation to efficiently utilize the scarce expert annotations, thus boosting the model's learning effectiveness. eCLIP is designed to be generally applicable to any variant of CLIP without requiring any modifications of the core architecture. Through detailed evaluations across several tasks, including zero-shot inference, linear probing, cross-modal retrieval, and Retrieval Augmented Generation (RAG) of radiology reports using a frozen Large Language Model, eCLIP showcases consistent improvements in embedding quality. The outcomes reveal enhanced alignment and uniformity, affirming eCLIP's capability to harness high-quality annotations for enriched multi-modal analysis in the medical imaging domain.

Deconfounded Representation Similarity for Comparison of Neural Networks

Similarity metrics such as representational similarity analysis (RSA) and centered kernel alignment (CKA) have been used to compare layer-wise representations between neural networks. However, these metrics are confounded by the population structure of data items in the input space, leading to spuriously high similarity for even completely random neural networks and inconsistent domain relations in transfer learning. We introduce a simple and generally applicable fix to adjust for the confounder with covariate adjustment regression, which retains the intuitive invariance properties of the original similarity measures. We show that deconfounding the similarity metrics increases the resolution of detecting semantically similar neural networks. Moreover, in real-world applications, deconfounding improves the consistency of representation similarities with domain similarities in transfer learning, and increases correlation with out-of-distribution accuracy.

Medical SANSformers: Training self-supervised transformers without attention for Electronic Medical Records

We leverage deep sequential models to tackle the problem of predicting healthcare utilization for patients, which could help governments to better allocate resources for future healthcare use. Specifically, we study the problem of \textit{divergent subgroups}, wherein the outcome distribution in a smaller subset of the population considerably deviates from that of the general population. The traditional approach for building specialized models for divergent subgroups could be problematic if the size of the subgroup is very small (for example, rare diseases). To address this challenge, we first develop a novel attention-free sequential model, SANSformers, instilled with inductive biases suited for modeling clinical codes in electronic medical records. We then design a task-specific self-supervision objective and demonstrate its effectiveness, particularly in scarce data settings, by pre-training each model on the entire health registry (with close to one million patients) before fine-tuning for downstream tasks on the divergent subgroups. We compare the novel SANSformer architecture with the LSTM and Transformer models using two data sources and a multi-task learning objective that aids healthcare utilization prediction. Empirically, the attention-free SANSformer models perform consistently well across experiments, outperforming the baselines in most cases by at least $\sim 10$\%. Furthermore, the self-supervised pre-training boosts performance significantly throughout, for example by over $\sim 50$\% (and as high as $800$\%) on $R^2$ score when predicting the number of hospital visits.