This work addresses the need for a balanced approach between performance and efficiency in scalable production environments for visually-rich document understanding (VDU) tasks. Currently, there is a reliance on large document foundation models that offer advanced capabilities but come with a heavy computational burden. In this paper, we propose a multimodal early exit (EE) model design that incorporates various training strategies, exit layer types and placements. Our goal is to achieve a Pareto-optimal balance between predictive performance and efficiency for multimodal document image classification. Through a comprehensive set of experiments, we compare our approach with traditional exit policies and showcase an improved performance-efficiency trade-off. Our multimodal EE design preserves the model's predictive capabilities, enhancing both speed and latency. This is achieved through a reduction of over 20% in latency, while fully retaining the baseline accuracy. This research represents the first exploration of multimodal EE design within the VDU community, highlighting as well the effectiveness of calibration in improving confidence scores for exiting at different layers. Overall, our findings contribute to practical VDU applications by enhancing both performance and efficiency.
While extensively explored in text-based tasks, Named Entity Recognition (NER) remains largely neglected in spoken language understanding. Existing resources are limited to a single, English-only dataset. This paper addresses this gap by introducing MSNER, a freely available, multilingual speech corpus annotated with named entities. It provides annotations to the VoxPopuli dataset in four languages (Dutch, French, German, and Spanish). We have also releasing an efficient annotation tool that leverages automatic pre-annotations for faster manual refinement. This results in 590 and 15 hours of silver-annotated speech for training and validation, alongside a 17-hour, manually-annotated evaluation set. We further provide an analysis comparing silver and gold annotations. Finally, we present baseline NER models to stimulate further research on this newly available dataset.
Argument structure learning~(ASL) entails predicting relations between arguments. Because it can structure a document to facilitate its understanding, it has been widely applied in many fields~(medical, commercial, and scientific domains). Despite its broad utilization, ASL remains a challenging task because it involves examining the complex relationships between the sentences in a potentially unstructured discourse. To resolve this problem, we have developed a simple yet effective approach called Dual-tower Multi-scale cOnvolution neural Network~(DMON) for the ASL task. Specifically, we organize arguments into a relationship matrix that together with the argument embeddings forms a relationship tensor and design a mechanism to capture relations with contextual arguments. Experimental results on three different-domain argument mining datasets demonstrate that our framework outperforms state-of-the-art models. The code is available at https://github.com/VRCMF/DMON.git .
Text-based semantic image editing assumes the manipulation of an image using a natural language instruction. Although recent works are capable of generating creative and qualitative images, the problem is still mostly approached as a black box sensitive to generating unexpected outputs. Therefore, we propose a novel model to enhance the text-based control of an image editor by explicitly reasoning about which parts of the image to alter or preserve. It relies on word alignments between a description of the original source image and the instruction that reflects the needed updates, and the input image. The proposed Diffusion Masking with word Alignments (DM-Align) allows the editing of an image in a transparent and explainable way. It is evaluated on a subset of the Bison dataset and a self-defined dataset dubbed Dream. When comparing to state-of-the-art baselines, quantitative and qualitative results show that DM-Align has superior performance in image editing conditioned on language instructions, well preserves the background of the image and can better cope with long text instructions.
Current diffusion models create photorealistic images given a text prompt as input but struggle to correctly bind attributes mentioned in the text to the right objects in the image. This is evidenced by our novel image-graph alignment model called EPViT (Edge Prediction Vision Transformer) for the evaluation of image-text alignment. To alleviate the above problem, we propose focused cross-attention (FCA) that controls the visual attention maps by syntactic constraints found in the input sentence. Additionally, the syntax structure of the prompt helps to disentangle the multimodal CLIP embeddings that are commonly used in T2I generation. The resulting DisCLIP embeddings and FCA are easily integrated in state-of-the-art diffusion models without additional training of these models. We show substantial improvements in T2I generation and especially its attribute-object binding on several datasets.\footnote{Code and data will be made available upon acceptance.
By training to predict the next token in an unlabeled corpus, large language models learn to perform many tasks without any labeled data. However, their next-token-prediction objective arguably limits their performance in scenarios that require planning, such as writing a coherent article. In this paper, we train a module for planning the future writing process via a self-supervised learning objective. By conditioning on generated latent plans, our model extends the successful language model formula to more abstract planning in an unsupervised way. Empirically, we demonstrate that our method improves language modeling performance in general, particularly with respect to the text structure. Because our framework uses a planner module that is unsupervised and external to the language model, new planner modules can be trained at large scale and easily be shared with the community.
Differentiating relationships between entity pairs with limited labeled instances poses a significant challenge in few-shot relation classification. Representations of textual data extract rich information spanning the domain, entities, and relations. In this paper, we introduce a novel approach to enhance information extraction combining multiple sentence representations and contrastive learning. While representations in relation classification are commonly extracted using entity marker tokens, we argue that substantial information within the internal model representations remains untapped. To address this, we propose aligning multiple sentence representations, such as the [CLS] token, the [MASK] token used in prompting, and entity marker tokens. Our method employs contrastive learning to extract complementary discriminative information from these individual representations. This is particularly relevant in low-resource settings where information is scarce. Leveraging multiple sentence representations is especially effective in distilling discriminative information for relation classification when additional information, like relation descriptions, are not available. We validate the adaptability of our approach, maintaining robust performance in scenarios that include relation descriptions, and showcasing its flexibility to adapt to different resource constraints.
Despite differing from the human language processing mechanism in implementation and algorithms, current language models demonstrate remarkable human-like or surpassing language capabilities. Should computational language models be employed in studying the brain, and if so, when and how? To delve into this topic, this paper reviews efforts in using computational models for brain research, highlighting emerging trends. To ensure a fair comparison, the paper evaluates various computational models using consistent metrics on the same dataset. Our analysis reveals that no single model outperforms others on all datasets, underscoring the need for rich testing datasets and rigid experimental control to draw robust conclusions in studies involving computational models.
In recent years, diffusion models have made remarkable strides in text-to-video generation, sparking a quest for enhanced control over video outputs to more accurately reflect user intentions. Traditional efforts predominantly focus on employing either semantic cues, like images or depth maps, or motion-based conditions, like moving sketches or object bounding boxes. Semantic inputs offer a rich scene context but lack detailed motion specificity; conversely, motion inputs provide precise trajectory information but miss the broader semantic narrative. For the first time, we integrate both semantic and motion cues within a diffusion model for video generation, as demonstrated in Fig 1. To this end, we introduce the Scene and Motion Conditional Diffusion (SMCD), a novel methodology for managing multimodal inputs. It incorporates a recognized motion conditioning module and investigates various approaches to integrate scene conditions, promoting synergy between different modalities. For model training, we separate the conditions for the two modalities, introducing a two-stage training pipeline. Experimental results demonstrate that our design significantly enhances video quality, motion precision, and semantic coherence.
Mainstream parameter-efficient fine-tuning (PEFT) methods, such as LoRA or Adapter, project a model's hidden states to a lower dimension, allowing pre-trained models to adapt to new data through this low-rank bottleneck. However, PEFT tasks involving multiple modalities, like vision-language (VL) tasks, require not only adaptation to new data but also learning the relationship between different modalities. Targeting at VL PEFT tasks, we propose a family of operations, called routing functions, to enhance VL alignment in the low-rank bottlenecks. The routing functions adopt linear operations and do not introduce new trainable parameters. In-depth analyses are conducted to study their behavior. In various VL PEFT settings, the routing functions significantly improve performance of the original PEFT methods, achieving over 20% improvement on VQAv2 ($\text{RoBERTa}_{\text{large}}$+ViT-L/16) and 30% on COCO Captioning (GPT2-medium+ViT-L/16). Also when fine-tuning a pre-trained multimodal model such as CLIP-BART, we observe smaller but consistent improvements across a range of VL PEFT tasks.