Jihwan Kim

Hideaki Takahashi, Jihwan Kim, Suman Jana et al.

Zero-knowledge (ZK) circuits enable privacy-preserving computations and are central to many cryptographic protocols. Systems like Circom simplify ZK development by combining witness computation and circuit constraints in one program. However, even small errors can compromise security of ZK programs -- under-constrained circuits may accept invalid witnesses, while over-constrained ones may reject valid ones. Static analyzers are often imprecise with high false positives, and formal tools struggle with real-world circuit scale. Additionally, existing tools overlook several critical behaviors, such as intermediate computations and program aborts, and thus miss many vulnerabilities. Our theoretical contribution is the Trace-Constraint Consistency Test (TCCT), a foundational, language-independent formulation of ZK circuit bugs. TCCT provides a unified semantics that subsumes prior definitions and captures both under- and over-constrained vulnerabilities, exposing the full space of ZK bugs that elude prior tools. Our systems contribution is zkFuzz, a novel program mutation-based fuzzing framework for detecting TCCT violations. zkFuzz systematically mutates the computational logic of Zk programs guided by a novel fitness function, and injects carefully crafted inputs using tailored heuristics to expose bugs. We evaluated zkFuzz on 452 real-world ZK circuits written in Circom, a leading programming system for ZK development. zkFuzz successfully identified 85 bugs, including 59 zero-days-39 of which were confirmed by developers and \nfixed fixed, including bugs undetectable by prior works due to their fundamentally limited formulations, earning thousands of bug bounties. Our preliminary research on Noir, another emerging DSL for ZK circuit, also demonstrates the feasibility of zkFuzz to support multiple DSLs.

Hyeongmin Lee, Jin-Young Kim, Kyungjune Baek et al.

In this work, we discuss evaluating video foundation models in a fair and robust manner. Unlike language or image foundation models, many video foundation models are evaluated with differing parameters (such as sampling rate, number of frames, pretraining steps, etc.), making fair and robust comparisons challenging. Therefore, we present a carefully designed evaluation framework for measuring two core capabilities of video comprehension: appearance and motion understanding. Our findings reveal that existing video foundation models, whether text-supervised like UMT or InternVideo2, or self-supervised like V-JEPA, exhibit limitations in at least one of these capabilities. As an alternative, we introduce TWLV-I, a new video foundation model that constructs robust visual representations for both motion- and appearance-based videos. Based on the average top-1 accuracy of linear probing on five action recognition benchmarks, pretrained only on publicly accessible datasets, our model shows a 4.6%p improvement compared to V-JEPA (ViT-L) and a 7.7%p improvement compared to UMT (ViT-L). Even when compared to much larger models, our model demonstrates a 7.2%p improvement compared to DFN (ViT-H), a 2.7%p improvement compared to V-JEPA (ViT-H) and a 2.8%p improvement compared to InternVideo2 (ViT-g). We provide embedding vectors obtained by TWLV-I from videos of several commonly used video benchmarks, along with evaluation source code that can directly utilize these embeddings. The code is available at https://github.com/twelvelabs-io/video-embeddings-evaluation-framework.

Kyudan Jung, Jihwan Kim, Soyoon Kim et al.

As the paradigm of AI shifts from text-based LLMs to Speech Language Models (SLMs), there is a growing demand for full-duplex systems capable of real-time, natural human-computer interaction. However, the development of such models is constrained by the scarcity of high-quality, multi-speaker conversational data, as existing large-scale resources are predominantly single-speaker or limited in volume. Addressing the complex dynamics of natural dialogue, such as overlapping and back-channeling remains a challenge, with standard processing pipelines suffering from diarization errors and ASR hallucinations. To bridge this gap, we present a robust and scalable open-source data processing pipeline designed for full-duplex model.

Jihwan Kim, Miso Lee, Jae-Pil Heo

Temporal Action Detection (TAD) is challenging but fundamental for real-world video applications. Recently, DETR-based models have been devised for TAD but have not performed well yet. In this paper, we point out the problem in the self-attention of DETR for TAD; the attention modules focus on a few key elements, called temporal collapse problem. It degrades the capability of the encoder and decoder since their self-attention modules play no role. To solve the problem, we propose a novel framework, Self-DETR, which utilizes cross-attention maps of the decoder to reactivate self-attention modules. We recover the relationship between encoder features by simple matrix multiplication of the cross-attention map and its transpose. Likewise, we also get the information within decoder queries. By guiding collapsed self-attention maps with the guidance map calculated, we settle down the temporal collapse of self-attention modules in the encoder and decoder. Our extensive experiments demonstrate that Self-DETR resolves the temporal collapse problem by keeping high diversity of attention over all layers.

Jihwan Kim, Youngdo Kim, Hyo Seung Lee et al.

Recent advances in deep learning-based image reconstruction techniques have led to significant progress in phase retrieval using digital in-line holographic microscopy (DIHM). However, existing deep learning-based phase retrieval methods have technical limitations in generalization performance and three-dimensional (3D) morphology reconstruction from a single-shot hologram of biological cells. In this study, we propose a novel deep learning model, named MorpHoloNet, for single-shot reconstruction of 3D morphology by integrating physics-driven and coordinate-based neural networks. By simulating the optical diffraction of coherent light through a 3D phase shift distribution, the proposed MorpHoloNet is optimized by minimizing the loss between the simulated and input holograms on the sensor plane. Compared to existing DIHM methods that face challenges with twin image and phase retrieval problems, MorpHoloNet enables direct reconstruction of 3D complex light field and 3D morphology of a test sample from its single-shot hologram without requiring multiple phase-shifted holograms or angle scanning. The performance of the proposed MorpHoloNet is validated by reconstructing 3D morphologies and refractive index distributions from synthetic holograms of ellipsoids and experimental holograms of biological cells. The proposed deep learning model is utilized to reconstruct spatiotemporal variations in 3D translational and rotational behaviors and morphological deformations of biological cells from consecutive single-shot holograms captured using DIHM. MorpHoloNet would pave the way for advancing label-free, real-time 3D imaging and dynamic analysis of biological cells under various cellular microenvironments in biomedical and engineering fields.

Won Ik Cho, Yoon Kyung Lee, Seoyeon Bae et al.

Building a natural language dataset requires caution since word semantics is vulnerable to subtle text change or the definition of the annotated concept. Such a tendency can be seen in generative tasks like question-answering and dialogue generation and also in tasks that create a categorization-based corpus, like topic classification or sentiment analysis. Open-domain conversations involve two or more crowdworkers freely conversing about any topic, and collecting such data is particularly difficult for two reasons: 1) the dataset should be ``crafted" rather than ``obtained" due to privacy concerns, and 2) paid creation of such dialogues may differ from how crowdworkers behave in real-world settings. In this study, we tackle these issues when creating a large-scale open-domain persona dialogue corpus, where persona implies that the conversation is performed by several actors with a fixed persona and user-side workers from an unspecified crowd.

Jihwan Kim, Miso Lee, Cheol-Ho Cho et al.

Temporal Action Detection (TAD) is fundamental yet challenging for real-world video applications. Leveraging the unique benefits of transformers, various DETR-based approaches have been adopted in TAD. However, it has recently been identified that the attention collapse in self-attention causes the performance degradation of DETR for TAD. Building upon previous research, this paper newly addresses the attention collapse problem in cross-attention within DETR-based TAD methods. Moreover, our findings reveal that cross-attention exhibits patterns distinct from predictions, indicating a short-cut phenomenon. To resolve this, we propose a new framework, Prediction-Feedback DETR (Pred-DETR), which utilizes predictions to restore the collapse and align the cross- and self-attention with predictions. Specifically, we devise novel prediction-feedback objectives using guidance from the relations of the predictions. As a result, Pred-DETR significantly alleviates the collapse and achieves state-of-the-art performance among DETR-based methods on various challenging benchmarks including THUMOS14, ActivityNet-v1.3, HACS, and FineAction.

Kyudan Jung, Jihwan Kim, Minwoo Lee et al.

Recent advancements in text-to-speech technologies enable generating high-fidelity synthetic speech nearly indistinguishable from real human voices. While recent studies show the efficacy of self-supervised learning-based speech encoders for deepfake detection, these models struggle to generalize across unseen speakers. Our quantitative analysis suggests these encoder representations are substantially influenced by speaker information, causing detectors to exploit speaker-specific correlations rather than artifact-related cues. We call this phenomenon speaker entanglement. To mitigate this reliance, we introduce SNAP, a speaker-nulling framework. We estimate a speaker subspace and apply orthogonal projection to suppress speaker-dependent components, isolating synthesis artifacts within the residual features. By reducing speaker entanglement, SNAP encourages detectors to focus on artifact-related patterns, leading to state-of-the-art performance.

Jihwan Kim, Nikhil Parthasarathy, Danfeng Qin et al.

The fundamental challenge in scaling Video Large Language Models (Video LLMs) to long-form video lies in managing the explosion of visual-token context length. Existing strategies predominantly focus on "post-hoc" token reduction -- reducing visual tokens after feature extraction to alleviate the LLM's computational overhead. While these methods effectively reduce the number of visual tokens, we observe that the primary latency bottleneck then shifts from the LLM to the expensive per-frame processing of the vision encoder. To address this, we introduce LiteFrame, a strong, yet highly efficient video encoder backbone for Video LLMs. To train LiteFrame, we propose Compressed Token Distillation (CTD), a novel training framework that teaches a compact student vision encoder to directly predict information-dense, spatio-temporally compressed representations produced by a large teacher vision model, effectively bypassing redundant computation. When coupled with further Language Model Adaptation (LMA), this approach results in a new latency-accuracy Pareto frontier -- compared with InternVL3-8B, LiteFrame provides a 35% reduction in end-to-end latency while processing 8$\times$ more frames and improves average video understanding accuracy across multiple benchmarks. Our results demonstrate a new potential path to unlocking longer-form video understanding under fixed compute budgets.

Yerim Jeon, Subeen Lee, Jihwan Kim et al.

Few-shot object counting has garnered significant attention for its practicality as it aims to count target objects in a query image based on given exemplars without additional training. However, the prevailing extract-and-match approach has a shortcoming: query and exemplar features lack interaction during feature extraction since they are extracted independently and later correlated based on similarity. This can lead to insufficient target awareness and confusion in identifying the actual target when multiple class objects coexist. To address this, we propose a novel framework, Mutually-Aware FEAture learning (MAFEA), which encodes query and exemplar features with mutual awareness from the outset. By encouraging interaction throughout the pipeline, we obtain target-aware features robust to a multi-category scenario. Furthermore, we introduce background token to effectively associate the query's target region with exemplars and decouple its background region. Our extensive experiments demonstrate that our model achieves state-of-the-art performance on FSCD-LVIS and FSC-147 benchmarks with remarkably reduced target confusion.

Jihwan Kim, Miso Lee, Jae-Pil Heo

Temporal action detection (TAD) is challenging, yet fundamental for real-world video applications. Recently, DETR-based models for TAD have been prevailing thanks to their unique benefits. However, transformers demand a huge dataset, and unfortunately data scarcity in TAD causes a severe degeneration. In this paper, we identify two crucial problems from data scarcity: attention collapse and imbalanced performance. To this end, we propose a new pre-training strategy, Long-Term Pre-training (LTP), tailored for transformers. LTP has two main components: 1) class-wise synthesis, 2) long-term pretext tasks. Firstly, we synthesize long-form video features by merging video snippets of a target class and non-target classes. They are analogous to untrimmed data used in TAD, despite being created from trimmed data. In addition, we devise two types of long-term pretext tasks to learn long-term dependency. They impose long-term conditions such as finding second-to-fourth or short-duration actions. Our extensive experiments show state-of-the-art performances in DETR-based methods on ActivityNet-v1.3 and THUMOS14 by a large margin. Moreover, we demonstrate that LTP significantly relieves the data scarcity issues in TAD.

Seungbin Yang, Jihwan Kim, Jaemin Choi et al.

Web information extraction (WIE) is the task of automatically extracting data from web pages, offering high utility for various applications. The evaluation of WIE systems has traditionally relied on benchmarks built from HTML snapshots captured at a single point in time. However, this offline evaluation paradigm fails to account for the temporally evolving nature of the web; consequently, performance on these static benchmarks often fails to generalize to dynamic real-world scenarios. To bridge this gap, we introduce \dataset, a new benchmark designed for evaluating WIE systems directly against live websites. Based on trusted and permission-granted websites, we curate natural language queries that require information extraction of various data categories, such as text, images, and hyperlinks. We further design these queries to represent four levels of complexity, based on the number and cardinality of attributes to be extracted, enabling a granular assessment of WIE systems. In addition, we propose Visual Grounding Scraper (VGS), a novel multi-stage agentic framework that mimics human cognitive processes by visually narrowing down web page content to extract desired information. Extensive experiments across diverse backbone models demonstrate the effectiveness and robustness of VGS. We believe that this study lays the foundation for developing practical and robust WIE systems.

Jihwan Kim, Jaehyun Choi, Yerim Jeon et al.

Temporal action detection (TAD) is challenging, yet fundamental for real-world video applications. Large temporal scale variation of actions is one of the most primary difficulties in TAD. Naturally, multi-scale features have potential in localizing actions of diverse lengths as widely used in object detection. Nevertheless, unlike objects in images, actions have more ambiguity in their boundaries. That is, small neighboring objects are not considered as a large one while short adjoining actions can be misunderstood as a long one. In the coarse-to-fine feature pyramid via pooling, these vague action boundaries can fade out, which we call 'vanishing boundary problem'. To this end, we propose Boundary-Recovering Network (BRN) to address the vanishing boundary problem. BRN constructs scale-time features by introducing a new axis called scale dimension by interpolating multi-scale features to the same temporal length. On top of scale-time features, scale-time blocks learn to exchange features across scale levels, which can effectively settle down the issue. Our extensive experiments demonstrate that our model outperforms the state-of-the-art on the two challenging benchmarks, ActivityNet-v1.3 and THUMOS14, with remarkably reduced degree of the vanishing boundary problem.

Taehee Kim, Seungbin Yang, Jihwan Kim et al.

Retrieving relevant tables from extensive databases for a given natural language query is essential for accurately answering questions in tasks such as text-to-SQL. Existing table retrieval approaches select a pre-determined set of k tables with the highest similarity to the query. However, the number of required tables varies across queries and cannot be known in advance. Enforcing a fixed number of retrieved tables regardless of the query may either retrieve an undersized set, failing to obtain all necessary evidence, or retrieve an oversized pool, including irrelevant tables. To address this issue, we propose an adaptive table retrieval method that adjusts the number of tables retrieved according to the requirements of each query. Specifically, we utilize an adaptive thresholding mechanism to selectively retrieve tables and integrate a sliding-window reranking algorithm to efficiently process a large table corpus. Extensive experiments on Spider, BIRD, and Spider 2.0 demonstrate that our method effectively addresses the limitations of the top-k retrieval strategy, improving performance in retrieval and downstream tasks. Our code and data are available at https://github.com/sbY99/Adaptive-Table-Retrieval.

Alex Mathai, Chenxi Huang, Suwei Ma et al.

Code large language models (LLMs) have shown impressive capabilities on a multitude of software engineering tasks. In particular, they have demonstrated remarkable utility in the task of code repair. However, common benchmarks used to evaluate the performance of code LLMs are often limited to small-scale settings. In this work, we build upon kGym, which shares a benchmark for system-level Linux kernel bugs and a platform to run experiments on the Linux kernel. This paper introduces CrashFixer, the first LLM-based software repair agent that is applicable to Linux kernel bugs. Inspired by the typical workflow of a kernel developer, we identify the key capabilities an expert developer leverages to resolve a kernel crash. Using this as our guide, we revisit the kGym platform and identify key system improvements needed to practically run LLM-based agents at the scale of the Linux kernel (50K files and 20M lines of code). We implement these changes by extending kGym to create an improved platform - called kGymSuite, which will be open-sourced. Finally, the paper presents an evaluation of various repair strategies for such complex kernel bugs and showcases the value of explicitly generating a hypothesis before attempting to fix bugs in complex systems such as the Linux kernel. We also evaluated CrashFixer's capabilities on still open bugs, and found at least two patch suggestions considered plausible to resolve the reported bug.

Jihwan Kim, Chenglin Fan

We study differentially private (DP) training with Muon, a matrix-valued optimizer that updates hidden-layer weights using momentum followed by Newton--Schulz orthogonalization. While DP-SGD is well understood, the interaction between per-example clipping, Gaussian noise, momentum, and nonlinear orthogonalization in Muon has not been systematically analyzed. We formulate DP-Muon, a private Muon procedure that clips per-example matrix gradients, adds Gaussian noise to the clipped lot average, and then applies momentum and Newton--Schulz orthogonalization as post-processing. We prove that DP-Muon inherits the privacy guarantee certified by the corresponding same-lot subsampled Gaussian accountant, with no additional privacy cost from Muon-specific post-processing. On the optimization side, we establish finite-horizon and vanishing stationarity guarantees under per-matrix clipping, with bounds that separate optimization error, clipping residual, privacy noise, and Newton--Schulz approximation error. We further show that the DP-induced bias in Muon arises not in the linear momentum buffer itself, but after the nonlinear Newton--Schulz map, where Gaussian noise induces a matrix-valued heat-smoothing bias. This motivates DP-MuonBC, a bias-corrected variant that removes the leading output-level bias term while preserving the same privacy guarantee. Experiments on E2E and DART show that Muon-style matrix updates improve private fine-tuning, and that DP-MuonBC further improves utility without increasing the privacy budget.

Jihwan Kim, Dogyoon Song, Chulhee Yun

We study scaling laws of signSGD under a power-law random features (PLRF) model that accounts for both feature and target decay. We analyze the population risk of a linear model trained with one-pass signSGD on Gaussian-sketched features. We express the risk as a function of model size, training steps, learning rate, and the feature and target decay parameters. Comparing against the SGD risk analyzed by Paquette et al. (2024), we identify a drift-normalization effect and a noise-reshaping effect unique to signSGD. We then obtain compute-optimal scaling laws under the optimal choice of learning rate. Our analysis shows that the noise-reshaping effect can make the compute-optimal slope of signSGD steeper than that of SGD in regimes where noise is dominant. Finally, we observe that the widely used warmup-stable-decay (WSD) schedule further reduces the noise term and sharpens the compute-optimal slope, when feature decay is fast but target decay is slow.

Jihwan Kim, Junoh Kang, Jinyoung Choi et al.

We propose a novel inference technique based on a pretrained diffusion model for text-conditional video generation. Our approach, called FIFO-Diffusion, is conceptually capable of generating infinitely long videos without additional training. This is achieved by iteratively performing diagonal denoising, which simultaneously processes a series of consecutive frames with increasing noise levels in a queue; our method dequeues a fully denoised frame at the head while enqueuing a new random noise frame at the tail. However, diagonal denoising is a double-edged sword as the frames near the tail can take advantage of cleaner frames by forward reference but such a strategy induces the discrepancy between training and inference. Hence, we introduce latent partitioning to reduce the training-inference gap and lookahead denoising to leverage the benefit of forward referencing. Practically, FIFO-Diffusion consumes a constant amount of memory regardless of the target video length given a baseline model, while well-suited for parallel inference on multiple GPUs. We have demonstrated the promising results and effectiveness of the proposed methods on existing text-to-video generation baselines. Generated video examples and source codes are available at our project page.

Jihwan Kim, Chenglin Fan

The ski rental problem is a canonical model for online decision-making under uncertainty, capturing the fundamental trade-off between repeated rental costs and a one-time purchase. While classical algorithms focus on worst-case competitive ratios and recent "learning-augmented" methods leverage point-estimate predictions, neither approach fully exploits the richness of full distributional predictions while maintaining rigorous robustness guarantees. We address this gap by establishing a systematic framework that integrates distributional advice of unknown quality into both deterministic and randomized algorithms. For the deterministic setting, we formalize the problem under perfect distributional prediction and derive an efficient algorithm to compute the optimal threshold-buy day. We provide a rigorous performance analysis, identifying sufficient conditions on the predicted distribution under which the expected competitive ratio (ECR) matches the classic optimal randomized bound. To handle imperfect predictions, we propose the Clamp Policy, which restricts the buying threshold to a safe range controlled by a tunable parameter. We show that this policy is both robust, maintaining good performance even with large prediction errors, and consistent, approaching the optimal performance as predictions become accurate. For the randomized setting, we characterize the stopping distribution via a Water-Filling Algorithm, which optimizes expected cost while strictly satisfying robustness constraints. Experimental results across diverse distributions (Gaussian, geometric, and bi-modal) demonstrate that our framework improves consistency significantly over existing point-prediction baselines while maintaining comparable robustness.

Miso Lee, Jihwan Kim, Jae-Pil Heo

Multi-scene absolute pose regression addresses the demand for fast and memory-efficient camera pose estimation across various real-world environments. Nowadays, transformer-based model has been devised to regress the camera pose directly in multi-scenes. Despite its potential, transformer encoders are underutilized due to the collapsed self-attention map, having low representation capacity. This work highlights the problem and investigates it from a new perspective: distortion of query-key embedding space. Based on the statistical analysis, we reveal that queries and keys are mapped in completely different spaces while only a few keys are blended into the query region. This leads to the collapse of the self-attention map as all queries are considered similar to those few keys. Therefore, we propose simple but effective solutions to activate self-attention. Concretely, we present an auxiliary loss that aligns queries and keys, preventing the distortion of query-key space and encouraging the model to find global relations by self-attention. In addition, the fixed sinusoidal positional encoding is adopted instead of undertrained learnable one to reflect appropriate positional clues into the inputs of self-attention. As a result, our approach resolves the aforementioned problem effectively, thus outperforming existing methods in both outdoor and indoor scenes.

Yeonsang Shin, Jihwan Kim, Yumin Song et al.

Despite the remarkable progress in text-to-video models, achieving precise control over text elements and animated graphics remains a significant challenge, especially in applications such as video advertisements. To address this limitation, we introduce Animated Layout Generation, a novel approach to extend static graphic layouts with temporal dynamics. We propose a Structured Text Representation for fine-grained video control through hierarchical visual elements. To demonstrate the effectiveness of our approach, we present VAKER (Video Ad maKER), a text-to-video advertisement generation pipeline that combines a three-stage generation process with Unstructured Text Reasoning for seamless integration with LLMs. VAKER fully automates video advertisement generation by incorporating dynamic layout trajectories for objects and graphics across specific video frames. Through extensive evaluations, we demonstrate that VAKER significantly outperforms existing methods in generating video advertisements. Project Page: https://yeonsangshin.github.io/projects/Vaker

Junghwan Lim, Gangwon Jo, Sungmin Lee et al.

We introduce Llama-3-Motif, a language model consisting of 102 billion parameters, specifically designed to enhance Korean capabilities while retaining strong performance in English. Developed on the Llama 3 architecture, Llama-3-Motif employs advanced training techniques, including LlamaPro and Masked Structure Growth, to effectively scale the model without altering its core Transformer architecture. Using the MoAI platform for efficient training across hyperscale GPU clusters, we optimized Llama-3-Motif using a carefully curated dataset that maintains a balanced ratio of Korean and English data. Llama-3-Motif shows decent performance on Korean-specific benchmarks, outperforming existing models and achieving results comparable to GPT-4.

Junghwan Lim, Sungmin Lee, Dongseok Kim et al.

Recent advancements in Large Language Models (LLMs) have revolutionized artificial intelligence, yet developing an effective foundational LLM that balances high performance with computational efficiency remains challenging, especially for emerging research groups. To address this gap, we introduce Motif-2.6B, a 2.6-billion-parameter foundation model designed to democratize advanced LLM capabilities. Motif-2.6B incorporates several innovative architectural enhancements, including Differential Attention and PolyNorm activation functions, which improve long-context comprehension, reduce hallucination, and enhance in-context learning capabilities. We rigorously tested multiple novel architectural components through extensive experimentation to determine the optimal architecture for Motif-2.6B. Comprehensive evaluations demonstrate that Motif-2.6B consistently meets or exceeds the performance of similarly sized state-of-the-art models across diverse benchmarks, showcasing its effectiveness, scalability, and real-world applicability. Through detailed experiments and tailored techniques, Motif-2.6B significantly advances the landscape of efficient, scalable, and powerful foundational LLMs, offering valuable insights and a robust foundation for future research and deployment.

Sung Hwan Mun, Min Hyun Han, Dongjune Lee et al.

In this paper, we propose self-supervised speaker representation learning strategies, which comprise of a bootstrap equilibrium speaker representation learning in the front-end and an uncertainty-aware probabilistic speaker embedding training in the back-end. In the front-end stage, we learn the speaker representations via the bootstrap training scheme with the uniformity regularization term. In the back-end stage, the probabilistic speaker embeddings are estimated by maximizing the mutual likelihood score between the speech samples belonging to the same speaker, which provide not only speaker representations but also data uncertainty. Experimental results show that the proposed bootstrap equilibrium training strategy can effectively help learn the speaker representations and outperforms the conventional methods based on contrastive learning. Also, we demonstrate that the integrated two-stage framework further improves the speaker verification performance on the VoxCeleb1 test set in terms of EER and MinDCF.

Jisung Wang, Jihwan Kim, Sangki Kim et al.

As the character-based end-to-end automatic speech recognition (ASR) models evolve, the choice of acoustic modeling units becomes important. Since Korean is a fairly phonetic language and has a unique writing system with its own Korean alphabet, it's worth investigating modeling units for an end-to-end Korean ASR task. In this work, we introduce lexicon-free modeling units in Korean, and explore them using a hybrid CTC/Attention-based encoder-decoder model. Five lexicon-free units are investigated: Syllable-based Korean character (with English character for a code-switching task), Korean Jamo character (with English character), sub-word on syllable-based character (with sub-word in English), sub-word on Jamo character (with sub-words in English), and finally byte unit, which is a universal one across language. Experiments on Zeroth-Korean (51.6 hrs) and Medical Record (2530 hrs) are done for Korean and Korean-English code-switching ASR tasks, respectively. Sequence-to-sequence learning with sub-words based on Korean syllables (and sub-words in English) performs the best for both tasks without lexicon and an extra language model integration.