Kyuhong Shim

Jaehoon Kang, Yejin Lee, Kyuhong Shim

We propose GLASS, a framework for composable acoustic style control in zero-shot autoregressive text-to-speech (TTS) that learns controls from post-generation rewards rather than style labels. In zero-shot TTS, a speaker prompt often entangles speaker identity with prosodic attributes such as speaking rate and pitch, making it difficult to change style without changing the prompt itself. GLASS instead treats each acoustic attribute as a reward-defined control direction. For each control axis, GLASS freezes the TTS backbone and trains one lightweight LoRA adapter with Group Relative Policy Optimization (GRPO), using speech-token length and mean F0 as style rewards and WER as an intelligibility anchor. Because each control is represented as a LoRA weight update, independently trained adapters can be swapped, interpolated, and composed through linear LoRA arithmetic without retraining the backbone. Experiments on speaking rate and pitch control show targeted style shifts while preserving naturalness, speaker similarity, and intelligibility, and demonstrate smooth interpolation and multi-axis composition across independently trained adapters.

Jiseob Kim, Kyuhong Shim, Junhan Kim et al.

Generalized zero-shot learning (GZSL) is a technique to train a deep learning model to identify unseen classes using the image attribute. In this paper, we put forth a new GZSL approach exploiting Vision Transformer (ViT) to maximize the attribute-related information contained in the image feature. In ViT, the entire image region is processed without the degradation of the image resolution and the local image information is preserved in patch features. To fully enjoy these benefits of ViT, we exploit patch features as well as the CLS feature in extracting the attribute-related image feature. In particular, we propose a novel attention-based module, called attribute attention module (AAM), to aggregate the attribute-related information in patch features. In AAM, the correlation between each patch feature and the synthetic image attribute is used as the importance weight for each patch. From extensive experiments on benchmark datasets, we demonstrate that the proposed technique outperforms the state-of-the-art GZSL approaches by a large margin.

Kyuhong Shim, Jiyoung Kim, Gusang Lee et al.

Monocular depth estimation is very challenging because clues to the exact depth are incomplete in a single RGB image. To overcome the limitation, deep neural networks rely on various visual hints such as size, shade, and texture extracted from RGB information. However, we observe that if such hints are overly exploited, the network can be biased on RGB information without considering the comprehensive view. We propose a novel depth estimation model named RElative Depth Transformer (RED-T) that uses relative depth as guidance in self-attention. Specifically, the model assigns high attention weights to pixels of close depth and low attention weights to pixels of distant depth. As a result, the features of similar depth can become more likely to each other and thus less prone to misused visual hints. We show that the proposed model achieves competitive results in monocular depth estimation benchmarks and is less biased to RGB information. In addition, we propose a novel monocular depth estimation benchmark that limits the observable depth range during training in order to evaluate the robustness of the model for unseen depths.

Sunwoo Kim, Kyuhong Shim, Luong Trung Nguyen et al.

Image text retrieval is a task to search for the proper textual descriptions of the visual world and vice versa. One challenge of this task is the vulnerability to input image and text corruptions. Such corruptions are often unobserved during the training, and degrade the retrieval model decision quality substantially. In this paper, we propose a novel image text retrieval technique, referred to as robust visual semantic embedding (RVSE), which consists of novel image-based and text-based augmentation techniques called semantic preserving augmentation for image (SPAugI) and text (SPAugT). Since SPAugI and SPAugT change the original data in a way that its semantic information is preserved, we enforce the feature extractors to generate semantic aware embedding vectors regardless of the corruption, improving the model robustness significantly. From extensive experiments using benchmark datasets, we show that RVSE outperforms conventional retrieval schemes in terms of image-text retrieval performance.

Wooje Park, Insu Lee, Soohyun Kim et al.

Large vision-language models (LVLMs) are increasingly being applied to multi-view image inputs captured from diverse viewpoints. However, despite this growing use, current LVLMs often confuse or mismatch visual information originating from different instances or viewpoints, a phenomenon we term multi-view hallucination. To systematically analyze this problem, we construct MVH-Bench, a benchmark comprising 4.8k question-answer pairs targeting two types of hallucination: cross-instance and cross-view. Empirical results show that recent LVLMs struggle to correctly associate visual evidence with its corresponding instance or viewpoint. To overcome this limitation, we propose Reference Shift Contrastive Decoding (RSCD), a training-free decoding technique that suppresses visual interference by generating negative logits through attention masking. Experiments on MVH-Bench with Qwen2.5-VL and LLaVA-OneVision demonstrate that RSCD consistently improves performance by up to 21.1 and 34.6 points over existing hallucination mitigation methods, highlighting the effectiveness of our approach.

Minsoo Kim, Kyuhong Shim, Seongmin Park et al.

Pre-trained Transformer models such as BERT have shown great success in a wide range of applications, but at the cost of substantial increases in model complexity. Quantization-aware training (QAT) is a promising method to lower the implementation cost and energy consumption. However, aggressive quantization below 2-bit causes considerable accuracy degradation due to unstable convergence, especially when the downstream dataset is not abundant. This work proposes a proactive knowledge distillation method called Teacher Intervention (TI) for fast converging QAT of ultra-low precision pre-trained Transformers. TI intervenes layer-wise signal propagation with the intact signal from the teacher to remove the interference of propagated quantization errors, smoothing loss surface of QAT and expediting the convergence. Furthermore, we propose a gradual intervention mechanism to stabilize the recovery of subsections of Transformer layers from quantization. The proposed schemes enable fast convergence of QAT and improve the model accuracy regardless of the diverse characteristics of downstream fine-tuning tasks. We demonstrate that TI consistently achieves superior accuracy with significantly lower fine-tuning iterations on well-known Transformers of natural language processing as well as computer vision compared to the state-of-the-art QAT methods.

Kyuhong Shim, Jungwook Choi, Wonyong Sung

Transformer-based deep neural networks have achieved great success in various sequence applications due to their powerful ability to model long-range dependency. The key module of Transformer is self-attention (SA) which extracts features from the entire sequence regardless of the distance between positions. Although SA helps Transformer performs particularly well on long-range tasks, SA requires quadratic computation and memory complexity with the input sequence length. Recently, attention map reuse, which groups multiple SA layers to share one attention map, has been proposed and achieved significant speedup for speech recognition models. In this paper, we provide a comprehensive study on attention map reuse focusing on its ability to accelerate inference. We compare the method with other SA compression techniques and conduct a breakdown analysis of its advantages for a long sequence. We demonstrate the effectiveness of attention map reuse by measuring the latency on both CPU and GPU platforms.

Kyuhong Shim, Wonyong Sung

Transformer-based speech recognition models have achieved great success due to the self-attention (SA) mechanism that utilizes every frame in the feature extraction process. Especially, SA heads in lower layers capture various phonetic characteristics by the query-key dot product, which is designed to compute the pairwise relationship between frames. In this paper, we propose a variant of SA to extract more representative phonetic features. The proposed phonetic self-attention (phSA) is composed of two different types of phonetic attention; one is similarity-based and the other is content-based. In short, similarity-based attention captures the correlation between frames while content-based attention only considers each frame without being affected by other frames. We identify which parts of the original dot product equation are related to two different attention patterns and improve each part with simple modifications. Our experiments on phoneme classification and speech recognition show that replacing SA with phSA for lower layers improves the recognition performance without increasing the latency and the parameter size.

Kyuhong Shim, Wonyong Sung

Phoneme recognition is a very important part of speech recognition that requires the ability to extract phonetic features from multiple frames. In this paper, we compare and analyze CNN, RNN, Transformer, and Conformer models using phoneme recognition. For CNN, the ContextNet model is used for the experiments. First, we compare the accuracy of various architectures under different constraints, such as the receptive field length, parameter size, and layer depth. Second, we interpret the performance difference of these models, especially when the observable sequence length varies. Our analyses show that Transformer and Conformer models benefit from the long-range accessibility of self-attention through input frames.

Kyuhong Shim, Jinkyu Lee, Simyung Chang et al.

Streaming automatic speech recognition (ASR) models are restricted from accessing future context, which results in worse performance compared to the non-streaming models. To improve the performance of streaming ASR, knowledge distillation (KD) from the non-streaming to streaming model has been studied, mainly focusing on aligning the output token probabilities. In this paper, we propose a layer-to-layer KD from the teacher encoder to the student encoder. To ensure that features are extracted using the same context, we insert auxiliary non-streaming branches to the student and perform KD from the non-streaming teacher layer to the non-streaming auxiliary layer. We design a special KD loss that leverages the autoregressive predictive coding (APC) mechanism to encourage the streaming model to predict unseen future contexts. Experimental results show that the proposed method can significantly reduce the word error rate compared to previous token probability distillation methods.

Joonhyeok Shin, Jaehoon Kang, Yujun Lee et al.

Selecting an appropriate background music (BGM) that supports natural human conversation is a common production step in media and interactive systems. In this paper, we introduce dialogue-conditioned BGM recommendation, where a model should select non-intrusive, fitting music for a multi-turn conversation that often contains no music descriptors. To study this novel problem, we present DialBGM, a benchmark of 1,200 open-domain daily dialogues, each paired with four candidate music clips and annotated with human preference rankings. Rankings are determined by background suitability criteria, including contextual relevance, non-intrusiveness, and consistency. We evaluate a wide range of open-source and proprietary models, including audio-language models and multimodal LLMs, and show that current models fall far short of human judgments; no model exceeds 35% Hit@1 when selecting the top-ranked clip. DialBGM provides a standardized benchmark for developing discourse-aware methods for BGM selection and for evaluating both retrieval-based and generative models.

Seunghan Yang, Byeonggeun Kim, Kyuhong Shim et al.

Few-shot keyword spotting (FS-KWS) models usually require large-scale annotated datasets to generalize to unseen target keywords. However, existing KWS datasets are limited in scale and gathering keyword-like labeled data is costly undertaking. To mitigate this issue, we propose a framework that uses easily collectible, unlabeled reading speech data as an auxiliary source. Self-supervised learning has been widely adopted for learning representations from unlabeled data; however, it is known to be suitable for large models with enough capacity and is not practical for training a small footprint FS-KWS model. Instead, we automatically annotate and filter the data to construct a keyword-like dataset, LibriWord, enabling supervision on auxiliary data. We then adopt multi-task learning that helps the model to enhance the representation power from out-of-domain auxiliary data. Our method notably improves the performance over competitive methods in the FS-KWS benchmark.

Minsoo Kim, Kyuhong Shim, Jungwook Choi et al.

Modern multimodal large language models (MLLMs) can reason over hour-long video, yet their key-value (KV) cache grows linearly with time-quickly exceeding the fixed memory of phones, AR glasses, and edge robots. Prior compression schemes either assume the whole video and user query are available offline or must first build the full cache, so memory still scales with stream length. InfiniPot-V is the first training-free, query-agnostic framework that enforces a hard, length-independent memory cap for streaming video understanding. During video encoding it monitors the cache and, once a user-set threshold is reached, runs a lightweight compression pass that (i) removes temporally redundant tokens via Temporal-axis Redundancy (TaR) metric and (ii) keeps semantically significant tokens via Value-Norm (VaN) ranking. Across four open-source MLLMs and four long-video and streaming-video benchmarks, InfiniPot-V cuts peak GPU memory by up to 94%, sustains real-time generation, and matches or surpasses full-cache accuracy-even in multi-turn dialogues. By dissolving the KV cache bottleneck without retraining or query knowledge, InfiniPot-V closes the gap for on-device streaming video assistants.

Insu Lee, Wooje Park, Jaeyun Jang et al.

Large vision-language models (LVLMs) are increasingly deployed in interactive applications such as virtual and augmented reality, where a first-person (egocentric) view captured by head-mounted cameras serves as key input. While this view offers fine-grained cues about user attention and hand-object interactions, its narrow field of view and lack of global context often lead to failures on spatially or contextually demanding queries. To address this, we introduce a framework that augments egocentric inputs with third-person (exocentric) views, providing complementary information such as global scene layout and object visibility to LVLMs. We present E3VQA, the first benchmark for multi-view question answering with 4K high-quality question-answer pairs grounded in synchronized ego-exo image pairs. Additionally, we propose M3CoT, a training-free prompting technique that constructs a unified scene representation by integrating scene graphs from three complementary perspectives. M3CoT enables LVLMs to reason more effectively across views, yielding consistent performance gains (4.84% for GPT-4o and 5.94% for Gemini 2.0 Flash) over a recent CoT baseline. Our extensive evaluation reveals key strengths and limitations of LVLMs in multi-view reasoning and highlights the value of leveraging both egocentric and exocentric inputs. The dataset and source code are available at https://github.com/Leeinsu1/Towards-Comprehensive-Scene-Understanding.

Hansol Park, Hoseong Ahn, Junwon Moon et al.

Hallucinations in vision-language models have been extensively studied using benchmarks that probe reliability in image-text settings. In contrast, the effect of spoken queries on multimodal hallucinations remains largely unexplored, despite the growing role of voice-driven interfaces. In this work, we investigate how spoken input influences hallucinations in multimodal large language models. We present RePOPE-Spk, an audio-augmented extension of the RePOPE benchmark, where queries are provided as speech under diverse acoustic conditions. Using RePOPE-Spk, we systematically evaluate both proprietary and open-source models. Experimental results show that hallucinations escalate when queries are spoken rather than written: error rates increase by 3% under clean speech and by up to 20% with environmental noise. Input order and query length further affect robustness, while strategies such as many-shot prompting and chain-of-thought reasoning offer partial but insufficient mitigation. These findings highlight a critical and underexplored challenge, opening new directions for building reliable voice interface systems.

Jihwan Bang, Seunghan Yang, Kyuhong Shim et al.

Retrieval-Augmented Generation (RAG) systems are typically evaluated under static assumptions, despite being frequently corrected through user or expert feedback in deployment. Existing evaluation protocols focus on overall accuracy and fail to capture how systems adapt after feedback is introduced. We introduce feedback adaptation as a problem setting for RAG systems, which asks how effectively and how quickly corrective feedback propagates to future queries. To make this behavior measurable, we propose two evaluation axes: correction lag, which captures the delay between feedback provision and behavioral change, and post-feedback performance, which measures reliability on semantically related queries after feedback. Using these metrics, we show that training-based approaches exhibit a trade-off between delayed correction and reliable adaptation. We further propose PatchRAG, a minimal inference-time instantiation that incorporates feedback without retraining, demonstrating immediate correction and strong post-feedback generalization under the proposed evaluation. Our results highlight feedback adaptation as a previously overlooked dimension of RAG system behavior in interactive settings.

Jaehoon Kang, Yejin Lee, Yoonji Park et al.

While prompt-based text-to-speech (TTS) models enable natural language-driven speaking style control, they often provide limited fine-grained control and apply a single global style across an utterance. This restricts practical use cases that require continuous style attribute interpolation across utterances and time-varying style transitions within a single utterance. In this paper, we propose novel techniques to achieve both capabilities in existing prompt-based TTS models. For inter-utterance style interpolation, we compute direction vectors between contrastive style prompts in the embedding space and perform simple interpolation, enabling smooth transitions between style characteristics. For intra-utterance style transition, we first identify a strong attention bias toward early tokens in autoregressive TTS decoders, causing the initial audio realization to dominate subsequent generation. To mitigate this effect, we introduce KV-cache swapping and sliding-window attention masking. Experiments demonstrate that our proposed inter-utterance interpolation achieves a 99-100% success rate in gender conversion, up to 36 Hz pitch variation, and up to 1.6 syllables-per-second speed change. Our intra-utterance transition maintains a speaker similarity of 0.81-0.91 and achieves perceptual smoothness scores of 3.48-4.48.

Seungnyun Kim, Jihoon Moon, Jinhong Kim et al.

Recently, we are witnessing the remarkable progress and widespread adoption of sensing technologies in autonomous driving, robotics, and metaverse. Considering the rapid advancement of computer vision (CV) technology to analyze the sensing information, we anticipate a proliferation of wireless applications exploiting the sensing and CV technologies in 6G. In this article, we provide a holistic overview of the sensing and CV-aided wireless communications (SVWC) framework for 6G. By analyzing the high-resolution sensing information through the powerful CV techniques, SVWC can quickly and accurately understand the wireless environments and then perform the wireless tasks. To demonstrate the efficacy of SVWC, we design the whole process of SVWC including the sensing dataset collection, DL model training, and execution of realistic wireless tasks. From the numerical evaluations on 6G communication scenarios, we show that SVWC achieves considerable performance gains over the conventional 5G systems in terms of positioning accuracy, data rate, and access latency.

Hoseong Ahn, Jeongyun Chae, Yoonji Park et al.

Long-form speech recognition with large encoder-decoder models such as Whisper often exhibit hallucinations, repetition loops, and content omissions. These errors can accumulate and be further amplified when the previous segment's transcription is used as decoding context. We propose Whisper-CD, a training-free contrastive decoding framework that contrasts clean-audio logits against negative logits computed from three acoustically motivated perturbations: Gaussian noise injection, silence signal, and audio temporal shift. We aggregate these negatives via the log-sum-exp operator, building a unified multi-negative objective for token-by-token decoding. Across five English long-form benchmarks, Whisper-CD reduces WER by up to 24.3pp on CORAAL and shows 48% faster token generation throughput than beam search. Because Whisper-CD operates purely at inference time, it can be applied as a drop-in replacement to already-deployed Whisper systems without retraining.

Hanna Lee, Tan Dat Nguyen, Jaehoon Kang et al.

Recent decoder-only autoregressive text-to-speech (AR-TTS) models produce high-fidelity speech, but their memory and compute costs scale quadratically with sequence length due to full self-attention. In this paper, we propose WAND, Windowed Attention and Knowledge Distillation, a framework that adapts pretrained AR-TTS models to operate with constant computational and memory complexity. WAND separates the attention mechanism into two: persistent global attention over conditioning tokens and local sliding-window attention over generated tokens. To stabilize fine-tuning, we employ a curriculum learning strategy that progressively tightens the attention window. We further utilize knowledge distillation from a full-attention teacher to recover high-fidelity synthesis quality with high data efficiency. Evaluated on three modern AR-TTS models, WAND preserves the original quality while achieving up to 66.2% KV cache memory reduction and length-invariant, near-constant per-step latency.

Donghoon Kim, Gusang Lee, Kyuhong Shim et al.

Recently, we have observed that Large Multi-modal Models (LMMs) are revolutionizing the way machines interact with the world, unlocking new possibilities across various multi-modal applications. To adapt LMMs for downstream tasks, parameter-efficient fine-tuning (PEFT) which only trains additional prefix tokens or modules, has gained popularity. Nevertheless, there has been little analysis of how PEFT works in LMMs. In this paper, we delve into the strengths and weaknesses of each tuning strategy, shifting the focus from the efficiency typically associated with these approaches. We first discover that model parameter tuning methods such as LoRA and Adapters distort the feature representation space learned during pre-training and limit the full utilization of pre-trained knowledge. We also demonstrate that prefix-tuning excels at preserving the representation space, despite its lower performance on downstream tasks. These findings suggest a simple two-step PEFT strategy called Prefix-Tuned PEFT (PT-PEFT), which successively performs prefix-tuning and then PEFT (i.e., Adapter, LoRA), combines the benefits of both. Experimental results show that PT-PEFT not only improves performance in image captioning and visual question answering compared to vanilla PEFT methods but also helps preserve the representation space of the four pre-trained models.

Donghoon Kim, Minji Bae, Kyuhong Shim et al.

Text-to-image generative models like DALL-E and Stable Diffusion have revolutionized visual content creation across various applications, including advertising, personalized media, and design prototyping. However, crafting effective textual prompts to guide these models remains challenging, often requiring extensive trial and error. Existing prompt inversion approaches, such as soft and hard prompt techniques, are not so effective due to the limited interpretability and incoherent prompt generation. To address these issues, we propose Visually Guided Decoding (VGD), a gradient-free approach that leverages large language models (LLMs) and CLIP-based guidance to generate coherent and semantically aligned prompts. In essence, VGD utilizes the robust text generation capabilities of LLMs to produce human-readable prompts. Further, by employing CLIP scores to ensure alignment with user-specified visual concepts, VGD enhances the interpretability, generalization, and flexibility of prompt generation without the need for additional training. Our experiments demonstrate that VGD outperforms existing prompt inversion techniques in generating understandable and contextually relevant prompts, facilitating more intuitive and controllable interactions with text-to-image models.

Jiyoung Kim, Kyuhong Shim, Insu Lee et al.

Unsupervised semantic segmentation (USS) aims to discover and recognize meaningful categories without any labels. For a successful USS, two key abilities are required: 1) information compression and 2) clustering capability. Previous methods have relied on feature dimension reduction for information compression, however, this approach may hinder the process of clustering. In this paper, we propose a novel USS framework called Expand-and-Quantize Unsupervised Semantic Segmentation (EQUSS), which combines the benefits of high-dimensional spaces for better clustering and product quantization for effective information compression. Our extensive experiments demonstrate that EQUSS achieves state-of-the-art results on three standard benchmarks. In addition, we analyze the entropy of USS features, which is the first step towards understanding USS from the perspective of information theory.

Eunji Kim, Kyuhong Shim, Simyung Chang et al.

A text encoder within Vision-Language Models (VLMs) like CLIP plays a crucial role in translating textual input into an embedding space shared with images, thereby facilitating the interpretative analysis of vision tasks through natural language. Despite the varying significance of different textual elements within a sentence depending on the context, efforts to account for variation of importance in constructing text embeddings have been lacking. We propose a framework of Semantic Token Reweighting to build Interpretable text embeddings (SToRI), which incorporates controllability as well. SToRI refines the text encoding process in CLIP by differentially weighting semantic elements based on contextual importance, enabling finer control over emphasis responsive to data-driven insights and user preferences. The efficacy of SToRI is demonstrated through comprehensive experiments on few-shot image classification and image retrieval tailored to user preferences.

Juntae Lee, Jihwan Bang, Seunghan Yang et al.

Retrieval-augmented generation (RAG) with large language models (LLMs) is especially valuable in specialized domains, where precision is critical. To more specialize the LLMs into a target domain, domain-specific RAG has recently been developed by allowing the LLM to access the target domain early via finetuning. The domain-specific RAG makes more sense in resource-constrained environments like edge devices, as they should perform a specific task (e.g. personalization) reliably using only small-scale LLMs. While the domain-specific RAG is well-aligned with edge devices in this respect, it often relies on widely-used reasoning techniques like chain-of-thought (CoT). The reasoning step is useful to understand the given external knowledge, and yet it is computationally expensive and difficult for small-scale LLMs to learn it. Tackling this, we propose the Chain of Rank (CoR) which shifts the focus from intricate lengthy reasoning to simple ranking of the reliability of input external documents. Then, CoR reduces computational complexity while maintaining high accuracy, making it particularly suited for resource-constrained environments. We attain the state-of-the-art (SOTA) results in benchmarks, and analyze its efficacy.

Insu Lee, Jiseob Kim, Kyuhong Shim et al.

Compositional Zero-Shot Learning (CZSL) aims to identify unseen state-object compositions by leveraging knowledge learned from seen compositions. Existing approaches often independently predict states and objects, overlooking their relationships. In this paper, we propose a novel framework, learning primitive relations (LPR), designed to probabilistically capture the relationships between states and objects. By employing the cross-attention mechanism, LPR considers the dependencies between states and objects, enabling the model to infer the likelihood of unseen compositions. Experimental results demonstrate that LPR outperforms state-of-the-art methods on all three CZSL benchmark datasets in both closed-world and open-world settings. Through qualitative analysis, we show that LPR leverages state-object relationships for unseen composition prediction.

Hyun Jong Yang, Howon Lee, Kyuhong Shim et al.

Imagine advanced humanoid robots, powered by multimodal large language models (MLLMs), coordinating missions across industries like warehouse logistics, manufacturing, and safety rescue. While individual robots show local autonomy, realistic tasks demand coordination among multiple agents sharing vast streams of sensor data. Communication is indispensable, yet transmitting comprehensive data can overwhelm networks, especially when a system-level orchestrator or cloud-based MLLM fuses multimodal inputs for route planning or anomaly detection. These tasks are often initiated by high-level natural language instructions. This intent serves as a filter for resource optimization: by understanding the goal via MLLMs, the system can selectively activate relevant sensing modalities, dynamically allocate bandwidth, and determine computation placement. Thus, R2X is fundamentally an intent-to-resource orchestration problem where sensing, communication, and computation are jointly optimized to maximize task-level success under resource constraints. This survey examines how integrated design paves the way for multi-robot coordination under MLLM guidance. We review state-of-the-art sensing modalities, communication strategies, and computing approaches, highlighting how reasoning is split between on-device models and powerful edge/cloud servers. We present four end-to-end demonstrations (sense -> communicate -> compute -> act): (i) digital-twin warehouse navigation with predictive link context, (ii) mobility-driven proactive MCS control, (iii) a FollowMe robot with a semantic-sensing switch, and (iv) real-hardware open-vocabulary trash sorting via edge-assisted MLLM grounding. We emphasize system-level metrics -- payload, latency, and success -- to show why R2X orchestration outperforms purely on-device baselines.

Junwon Moon, Hyunjin Choi, Hansol Park et al.

Target speaker extraction (TSE) extracts the target speaker's voice from overlapping speech mixtures given a reference utterance. Existing approaches typically fall into two categories: discriminative and generative. Discriminative methods apply time-frequency masking for fast inference but often over-suppress the target signal, while generative methods synthesize high-quality speech at the cost of numerous iterative steps. We propose Mask2Flow-TSE, a two-stage framework combining the strengths of both paradigms. The first stage applies discriminative masking for coarse separation, and the second stage employs flow matching to refine the output toward target speech. Unlike generative approaches that synthesize speech from Gaussian noise, our method starts from the masked spectrogram, enabling high-quality reconstruction in a single inference step. Experiments show that Mask2Flow-TSE achieves comparable performance to existing generative TSE methods with approximately 85M parameters.

Donghoon Kim, Minji Bae, Unghui Nam et al.

Vision language action models (VLAs) are increasingly used for Physical AI, but deploying a pre-trained VLA model to unseen environments, embodiments, or tasks still requires adaptation. Parameter-efficient fine-tuning (PEFT), especially LoRA, is common for VLA policies, yet the exposed capacity knob, the rank, does not transfer uniformly: robotics transfer exhibits a higher and task-varying intrinsic rank than language fine-tuning. Small ranks suffice for LLMs (e.g., $r \in \{4, 8\}$), while spectral analyses indicate VLAs may require much larger ranks (e.g., $r \approx 128$) or near-full rank, a mismatch that worsens in multi-task settings. We present LoRA-SP (Select-Prune), a rank-adaptive fine-tuning method that replaces fixed-rank updates with input- and layer-wise capacity. LoRA-SP uses an SVD-style parameterization with a small router whose nonnegative scores act as singular values over a shared vector bank. The active set is chosen by an energy target on the cumulative squared scores $E(k) \ge η$, providing a direct link to approximation error via our spectral analysis. During training, $η$ concentrates energy on a few directions and teaches the router to rely on fewer vectors while preserving accuracy. This yields compact adapters that reduce cross-task interference and improve generalization. On four real-robot manipulation tasks collected on an unseen AgileX PiPER arm, across two VLA backbones ($π_0$ and SmolVLA), LoRA-SP matches or exceeds full fine-tuning with far fewer trainable parameters, and improves multi-task success by up to 31.6% over standard LoRA while remaining robust to rank choice.

Yejin Lee, Jaehoon Kang, Kyuhong Shim

While persona-driven large language models (LLMs) and prompt-based text-to-speech (TTS) systems have advanced significantly, a usability gap arises when users attempt to generate voices matching their desired personas from implicit descriptions. Most users lack specialized knowledge to specify detailed voice attributes, which often leads TTS systems to misinterpret their expectations. To address these gaps, we introduce Persona-to-Voice-Attribute (P2VA), the first framework enabling voice generation automatically from persona descriptions. Our approach employs two strategies: P2VA-C for structured voice attributes, and P2VA-O for richer style descriptions. Evaluation shows our P2VA-C reduces WER by 5% and improves MOS by 0.33 points. To the best of our knowledge, P2VA is the first framework to establish a connection between persona and voice synthesis. In addition, we discover that current LLMs embed societal biases in voice attributes during the conversion process. Our experiments and findings further provide insights into the challenges of building persona-voice systems.

Byeonggeun Kim, Juntae Lee, Kyuhong Shim et al.

Few-Shot Open-Set Recognition (FSOSR) targets a critical real-world challenge, aiming to categorize inputs into known categories, termed closed-set classes, while identifying open-set inputs that fall outside these classes. Although transfer learning where a model is tuned to a given few-shot task has become a prominent paradigm in closed-world, we observe that it fails to expand to open-world. To unlock this challenge, we propose a two-stage method which consists of open-set aware meta-learning with open-set free transfer learning. In the open-set aware meta-learning stage, a model is trained to establish a metric space that serves as a beneficial starting point for the subsequent stage. During the open-set free transfer learning stage, the model is further adapted to a specific target task through transfer learning. Additionally, we introduce a strategy to simulate open-set examples by modifying the training dataset or generating pseudo open-set examples. The proposed method achieves state-of-the-art performance on two widely recognized benchmarks, miniImageNet and tieredImageNet, with only a 1.5\% increase in training effort. Our work demonstrates the effectiveness of transfer learning in FSOSR.

Jihwan Bang, Juntae Lee, Kyuhong Shim et al.

The customization of large language models (LLMs) for user-specified tasks gets important. However, maintaining all the customized LLMs on cloud servers incurs substantial memory and computational overheads, and uploading user data can also lead to privacy concerns. On-device LLMs can offer a promising solution by mitigating these issues. Yet, the performance of on-device LLMs is inherently constrained by the limitations of small-scaled models. To overcome these restrictions, we first propose Crayon, a novel approach for on-device LLM customization. Crayon begins by constructing a pool of diverse base adapters, and then we instantly blend them into a customized adapter without extra training. In addition, we develop a device-server hybrid inference strategy, which deftly allocates more demanding queries or non-customized tasks to a larger, more capable LLM on a server. This ensures optimal performance without sacrificing the benefits of on-device customization. We carefully craft a novel benchmark from multiple question-answer datasets, and show the efficacy of our method in the LLM customization.

Kyuhong Shim, Hyewon Bae, Wonyong Sung

The performance of automatic speech recognition (ASR) models can be greatly improved by proper beam-search decoding with external language model (LM). There has been an increasing interest in Korean speech recognition, but not many studies have been focused on the decoding procedure. In this paper, we propose a Korean tokenization method for neural network-based LM used for Korean ASR. Although the common approach is to use the same tokenization method for external LM as the ASR model, we show that it may not be the best choice for Korean. We propose a new tokenization method that inserts a special token, SkipTC, when there is no trailing consonant in a Korean syllable. By utilizing the proposed SkipTC token, the input sequence for LM becomes very regularly patterned so that the LM can better learn the linguistic characteristics. Our experiments show that the proposed approach achieves a lower word error rate compared to the same LM model without SkipTC. In addition, we are the first to report the ASR performance for the recently introduced large-scale 7,600h Korean speech dataset.

Junhan Kim, Kyuhong Shim, Byonghyo Shim

Generalized zero-shot learning (GZSL) is a technique to train a deep learning model to identify unseen classes using the attribute. In this paper, we put forth a new GZSL technique that improves the GZSL classification performance greatly. Key idea of the proposed approach, henceforth referred to as semantic feature extraction-based GZSL (SE-GZSL), is to use the semantic feature containing only attribute-related information in learning the relationship between the image and the attribute. In doing so, we can remove the interference, if any, caused by the attribute-irrelevant information contained in the image feature. To train a network extracting the semantic feature, we present two novel loss functions, 1) mutual information-based loss to capture all the attribute-related information in the image feature and 2) similarity-based loss to remove unwanted attribute-irrelevant information. From extensive experiments using various datasets, we show that the proposed SE-GZSL technique outperforms conventional GZSL approaches by a large margin.

Kyuhong Shim, Iksoo Choi, Wonyong Sung et al.

While Transformer-based models have shown impressive language modeling performance, the large computation cost is often prohibitive for practical use. Attention head pruning, which removes unnecessary attention heads in the multihead attention, is a promising technique to solve this problem. However, it does not evenly reduce the overall load because the heavy feedforward module is not affected by head pruning. In this paper, we apply layer-wise attention head pruning on All-attention Transformer so that the entire computation and the number of parameters can be reduced proportionally to the number of pruned heads. While the architecture has the potential to fully utilize head pruning, we propose three training methods that are especially helpful to minimize performance degradation and stabilize the pruning process. Our pruned model shows consistently lower perplexity within a comparable parameter size than Transformer-XL on WikiText-103 language modeling benchmark.