Min Soo Kim

Jiwan Chung, Junhyeok Kim, Siyeol Kim et al.

When thinking with images, humans rarely rely on a single glance: they revisit visual information repeatedly during reasoning. However, existing models typically process images only once and thereafter generate reasoning entirely in text, lacking mechanisms to re-access or ground inference in visual representations. We empirically confirm this: as reasoning chains lengthen, models progressively lose focus on relevant regions. In response, we introduce v1, a lightweight extension that enables active visual referencing through a simple point-and-copy approach. This allows the model to identify relevant image patches and copy their embeddings back into the reasoning stream, ensuring that evolving hypotheses remain grounded in perceptual evidence. Crucially, our pointing strategy lets the MLLM directly select image patches using their semantic representations as keys, keeping perceptual evidence embedded in the same space as the model's reasoning. To train this capability, we construct v1g, a dataset of 300K multimodal reasoning traces with interleaved visual grounding annotations. Across various multimodal mathematical reasoning benchmarks, v1 consistently outperforms comparable baselines, establishing point-and-copy as a practical mechanism for grounded reasoning. The model checkpoint and dataset are available at github.com/jun297/v1.

Junhyeok Kim, Min Soo Kim, Jiwan Chung et al.

Predicting when to initiate speech in real-world environments remains a fundamental challenge for conversational agents. We introduce EgoSpeak, a novel framework for real-time speech initiation prediction in egocentric streaming video. By modeling the conversation from the speaker's first-person viewpoint, EgoSpeak is tailored for human-like interactions in which a conversational agent must continuously observe its environment and dynamically decide when to talk. Our approach bridges the gap between simplified experimental setups and complex natural conversations by integrating four key capabilities: (1) first-person perspective, (2) RGB processing, (3) online processing, and (4) untrimmed video processing. We also present YT-Conversation, a diverse collection of in-the-wild conversational videos from YouTube, as a resource for large-scale pretraining. Experiments on EasyCom and Ego4D demonstrate that EgoSpeak outperforms random and silence-based baselines in real time. Our results also highlight the importance of multimodal input and context length in effectively deciding when to speak.

Jaehyun Jeon, Min Soo Kim, Jang Han Yoon et al.

User interface (UI) design goes beyond visuals, guiding user behavior and overall user experience (UX). Strategically crafted interfaces, for example, can boost sign-ups and drive business sales, underscoring the shift toward UI/UX as a unified design concept. While recent studies have explored UI quality evaluation using Multimodal Large Language Models (MLLMs), they largely focus on surface-level features, overlooking behavior-oriented aspects. To fill this gap, we introduce WiserUI-Bench, a novel benchmark for assessing models' multimodal understanding of UI/UX design. It includes 300 diverse real-world UI image pairs, each consisting of two design variants A/B-tested at scale by actual companies, where one was empirically validated to steer more user actions than the other. Each pair is accompanied one or more of 684 expert-curated rationales that capture key factors behind each winning design's effectiveness, spanning diverse cognitive dimensions of UX. Our benchmark supports two core tasks: (1) selecting the more effective UI/UX design by predicting the A/B test verified winner and (2) assessing how well a model, given the winner, can explain its effectiveness in alignment with expert reasoning. Experiments across several MLLMs show that current models exhibit limited nuanced reasoning about UI/UX design and its behavioral impact. We believe our work will foster research in UI/UX understanding and enable broader applications such as behavior-aware interface optimization.

Eun-Soo Jung, HyeongGwan Son, Kyusam Oh et al.

We present a novel deep neural model for text detection in document images. For robust text detection in noisy scanned documents, the advantages of multi-task learning are adopted by adding an auxiliary task of text enhancement. Namely, our proposed model is designed to perform noise reduction and text region enhancement as well as text detection. Moreover, we enrich the training data for the model with synthesized document images that are fully labeled for text detection and enhancement, thus overcome the insufficiency of labeled document image data. For the effective exploitation of the synthetic and real data, the training process is separated in two phases. The first phase is training only synthetic data in a fully-supervised manner. Then real data with only detection labels are added in the second phase. The enhancement task for the real data is weakly-supervised with information from their detection labels. Our methods are demonstrated in a real document dataset with performances exceeding those of other text detection methods. Moreover, ablations are conducted and the results confirm the effectiveness of the synthetic data, auxiliary task, and weak-supervision. Whereas the existing text detection studies mostly focus on the text in scenes, our proposed method is optimized to the applications for the text in scanned documents.

Raul Murillo, Alberto A. Del Barrio, Guillermo Botella et al.

The Posit Number System was introduced in 2017 as a replacement for floating-point numbers. Since then, the community has explored its application in Neural Network related tasks and produced some unit designs which are still far from being competitive with their floating-point counterparts. This paper proposes a Posit Logarithm-Approximate Multiplication (PLAM) scheme to significantly reduce the complexity of posit multipliers, the most power-hungry units within Deep Neural Network architectures. When comparing with state-of-the-art posit multipliers, experiments show that the proposed technique reduces the area, power, and delay of hardware multipliers up to 72.86%, 81.79%, and 17.01%, respectively, without accuracy degradation.

Min Soo Kim, Alberto A. Del Barrio, HyunJin Kim et al.

This paper analyzes the effects of approximate multiplication when performing inferences on deep convolutional neural networks (CNNs). The approximate multiplication can reduce the cost of the underlying circuits so that CNN inferences can be performed more efficiently in hardware accelerators. The study identifies the critical factors in the convolution, fully-connected, and batch normalization layers that allow more accurate CNN predictions despite the errors from approximate multiplication. The same factors also provide an arithmetic explanation of why bfloat16 multiplication performs well on CNNs. The experiments are performed with recognized network architectures to show that the approximate multipliers can produce predictions that are nearly as accurate as the FP32 references, without additional training. For example, the ResNet and Inception-v4 models with Mitch-$w$6 multiplication produces Top-5 errors that are within 0.2% compared to the FP32 references. A brief cost comparison of Mitch-$w$6 against bfloat16 is presented, where a MAC operation saves up to 80% of energy compared to the bfloat16 arithmetic. The most far-reaching contribution of this paper is the analytical justification that multiplications can be approximated while additions need to be exact in CNN MAC operations.