arXiv subjects

Satoshi Nakamura

Publications and source records attributed to Satoshi Nakamura.

At least 19 recordsLinked to original sources

arXiv2026

Rubric-Aligned Disentangled Evaluation of Human Simultaneous Interpreting

Human simultaneous interpreting (SI) is commonly assessed with analytic rubrics separating meaning transfer, delivery quality, and temporal synchrony, yet no automatic metric is designed for rubric-aligned segment-level SI evaluation. We construct a professionally annotated corpus of 1,101 SI segments with scores for meaning transfer (LQ), delivery quality (EXP), and perceived latency (LAT). We show that structured LLM prompting and scalar supervision collapse rubric dimensions, yielding near-zero correlation with human ratings and strong cross-dimension coupling. To isolate supervision structure under identical backbone capacity, we introduce dual regression heads on a LoRA-adapted COMET-KIWI encoder. On a held-out talk-level test set, the model achieves Pearson correlations of 0.388 (LQ) and 0.301 (EXP), improving over frozen COMET-KIWI. Given low absolute rater agreement, we interpret results relative to human consistency and target stable ranking signals for formative assessment.

cs.CL↗

arXiv2026

When to Use Extra Context: Evidence-Grounded Terminology Adaptation for Simultaneous Speech Translation

Extra context is valuable for simultaneous speech translation of technical talks, but injecting the entire document context into every streaming segment is often too coarse. Through diagnostic experiments, we find that context gains mainly come from paper-specific terminology recovery rather than uniform semantic enhancement. We therefore propose EGTA, an Evidence-Grounded Terminology Adaptation framework that builds a document terminology memory, selects compact candidate terms conditioned on the current streaming state, and adapts ASR/speech-side and decoder-side decision spaces using only the selected terms. EGTA can be instantiated in cascaded, end-to-end, and generation-only SimulST settings without full-model fine-tuning. We evaluate EGTA on an ACL technical-talk SimulST evaluation suite consisting of MCIF-dev and ACL60/60-dev. On MCIF-dev, EGTA-RG improves BLEU by +1.05/+0.59, XCOMET-XL by +0.019/+0.006, named-entity recall by +79\%/+73\% relative, and acronym recall by +0.099/+0.171 on En$\rightarrow$Zh and En$\rightarrow$De. Across MCIF-dev latency settings, EGTA consistently improves XCOMET-XL, named-entity recall, and acronym recall. External validation on ACL60/60-dev further shows consistent terminology-recall gains without additional fine-tuning. Shuffled-memory controls and activation audits provide evidence that the improvements are tied to paper-specific evidence alignment rather than generic context prompting.

cs.CL↗

arXiv2026

Evaluating and Preserving Lexical Stress in English-to-Chinese Speech-to-Speech Translation

Speech-to-speech translation (S2ST) systems have achieved impressive progress in semantic accuracy and speech naturalness. However, the cross-lingual transfer of lexical stress, a vital cue for emphasis and speaker intent, remains heavily underexplored, compounded by a lack of reliable automatic evaluation metrics for tonal languages like Chinese. We investigate English-to-Chinese S2ST stress transfer by constructing a stress-annotated Chinese dataset and an XLS-R-based Mandarin stress detector. Integrating this with the English EmphAssess system, we propose a novel objective metric for cross-lingual stress evaluation. Furthermore, we fine-tune CosyVoice3 to build a stress-aware S2ST system. Experiments demonstrate that our proposed S2ST architecture significantly outperforms existing systems in stress translation capability while maintaining competitive translation quality. Furthermore, our evaluation metric exhibits a strong correlation with human subjective judgments.

cs.CL↗

arXiv2026

Leveraging Audio-LLMs to Filter Speech-to-Speech Training Data

Large-scale mined corpora provide abundant training data for end-to-end speech-to-speech translation (S2ST) but may contain noise, misalignment, and semantic errors. Filtering noisy data is crucial to maintain robust speech translation performance. We study how to train an audio-language model to make keep/drop decisions on paired speech directly from audio. To obtain reliable supervision without manual labels, we adopt a scalable two-stage Rank-to-Distill strategy. A lightweight ranker generates keep/drop pseudo-labels from noisy speech pairs, then trains an audio large language model to predict keep/drop directly from raw paired speech. The resulting model jointly captures acoustic fidelity and cross-lingual semantic consistency for the selection of speech-conditioned data. Experiments on CVSS-C and SpeechMatrix show consistent improvements over unfiltered training, yielding up to +1.4 ASR-BLEU for end-to-end S2ST.

cs.CL↗

arXiv2026

Gradient-Informed Training for Low-Resource Multilingual Speech Translation

In low-resource multilingual speech-to-text translation, uniform architectural sharing across languages frequently introduces representation conflicts that impede convergence. This work proposes a principled methodology to automatically determine layer-specific sharing patterns by mining training gradient information. Our approach employs three distinct analysis strategies: distance-based language clustering, self/cross-task divergence metrics for capacity allocation, and joint factorization coupled with canonical correlation analysis for subspace alignment. Extensive evaluation across four language pairs (using the SeamlessM4T-Medium architecture) demonstrates persistent improvements in translation quality metrics.

cs.CL↗

arXiv2026

Compact sub-10 ps Resolution Radio Frequency Photomultiplier Tube

Experimental measurements of the radial spreading of photoelectrons emitted from a multi-alkali photocathode in a MCP-based photomultiplier tube have shown that, for photon wavelengths of 455 nm, 515 nm and 625 nm, the maximum initial energies of the emitted photoelectrons are approximately 0.3 eV, 0.2 eV and 0.1 eV respectively. Combining these experimental results with simulations performed using the SIMION simulation package, a compact radio-frequency photoelectron multiplier tube with a temporal resolution better than 10 ps is proposed. The device would be suitable for applications in several fields, particularly in medical optical instruments employing time-correlated single-photon counting.

physics.ins-det↗

arXiv2026

Improving Data Quality via Pre-Task Participant Screening in Crowdsourced GUI Experiments

In crowdsourced user experiments that collect performance data from graphical user interface (GUI) interactions, some participants ignore instructions or act carelessly, threatening the validity of performance models. We investigate a pre-task screening method that requires simple GUI operations analogous to the main task and uses the resulting error as a continuous quality signal. Our pre-task is a brief image-resizing task in which workers match an on-screen card to a physical card; workers whose resizing error exceeds a threshold are excluded from the main experiment. The main task is a standardized pointing experiment with well-established models of movement time and error rate. Across mouse- and smartphone-based crowdsourced experiments, we show that reducing the proportion of workers exhibiting unexpected behavior and tightening the pre-task threshold systematically improve the goodness of fit and predictive accuracy of GUI performance models, demonstrating that brief pre-task screening can enhance data quality.

cs.HC↗

arXiv2026

Soliton-type metrics associated with weighted CSCK metrics on Fano manifolds

We study weighted constant scalar curvature K\"ahler metrics, introduced by Lahdili as $(v,w)$-CSCK metrics, on Fano manifolds and their relationship with soliton-type metrics. In this paper, we introduce a weight function $g(v,w)$ associated with a pair of weight functions $(v,w)$. Assuming that $v$ and $g(v,w)$ are positive and log-concave on the moment polytope, we prove that the existence of a $(v,w)$-CSCK metric in the first Chern class is equivalent to the existence of a $g(v,w)$-soliton. We also explain that a $g(v,w)$-soliton arises naturally from Sasaki geometry. More precisely, let $(v,w)$ be the weight functions defining a weighted CSCK metric in $2\pi c_1(X)$ which gives rise to a $\hat{\xi}$-transverse extremal metric on an $S^1$-bundle $N$ in the canonical bundle of a Fano manifold $X$, where $\hat{\xi}$ is a possibly irregular Reeb field on $N$. We prove that the associated $g(v,w)$-soliton on $X$ gives rise to a $\hat{\xi}$-transverse Mabuchi soliton on $N$.

math.DG↗

arXiv2026

Understanding Mental States to Guide Social Influence in Multi-Person Group Dialogue

Existing dynamic Theory of Mind (ToM) benchmarks mostly place language models in a passive role: the model reads a sequence of connected scenarios and reports what people believe, feel, intend, and do as these states change. In real social interaction, ToM is also used for action: a speaker plans what to say in order to shift another person's mental-state trajectory toward a goal. We introduce SocialMindChange, a benchmark that moves from tracking minds to changing minds in social interaction. Each instance defines a social context with 4 characters and five connected scenes. The model plays one character and generates dialogue across the five scenes to reach the target while remaining consistent with the evolving states of all participants. SocialMindChange also includes selected higher-order states. Using a structured four-step framework, we construct 1,200 social contexts, covering 6000 scenarios and over 90,000 questions, each validated for realism and quality. Evaluations on ten state-of-the-art LLMs show that their average performance is 54.2% below human performance. This gap suggests that current LLMs still struggle to maintain and change mental-state representations across long, linked interactions.

cs.AI↗

arXiv2026

Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies

Simultaneous Machine Translation (SiMT) requires high-quality translations under strict real-time constraints, which traditional policies with only READ/WRITE actions cannot fully address. We extend the action space of SiMT with four adaptive actions: Sentence_Cut, Drop, Partial_Summarization and Pronominalization, which enable real-time restructuring, omission, and simplification while preserving semantic fidelity. We adapt these actions in a large language model (LLM) framework and construct training references through action-aware prompting. To evaluate both quality and word-level monotonicity, we further develop a latency-aware TTS pipeline that maps textual outputs to speech with realistic timing. Experiments on the ACL60/60 English-Chinese, English-German and English-Japanese benchmarks show that our framework consistently improves semantic metrics and achieves lower delay compared to reference translations and salami-based baselines. Notably, combining Drop and Sentence_Cut leads to consistent improvements in the balance between fluency and latency. These results demonstrate that enriching the action space of LLM-based SiMT provides a promising direction for bridging the gap between human and machine interpretation.

cs.CL↗

arXiv2025

Picosecond Precision Heavy-Ion Detector for {\Lambda} Hypernuclei Lifetime Studies

In this paper, we present the design and preliminary performance evaluation of a new heavy-ion detector for direct measurements of heavy {\Lambda} hypernuclei lifetime. The detector employs the previously developed 10 picosecond resolution Radio Frequency (RF) Timer, which converts the temporal information of incident particles into spatial coordinates of secondary or photoelectrons on a position-sensitive detector by means of circular RF scanning in the 500-1000 MHz range. Here, we report the detector design to achieve efficient suppression of accidental background and effective separation of prompt reaction products and delayed events from {\Lambda} hypernuclei decays, results of test studies carried out with RF synchronized laser as well as preliminary results obtained by using alpha particles. Dedicated Monte-Carlo simulations have been performed to estimate the detector's performance under realistic experimental conditions at RF-driven electron, photon, or proton beams. The results confirm the feasibility of the proposed design and provide a basis for upcoming experimental measurements, based on the delayed fission detection.

physics.ins-det↗

arXiv2025

StressTransfer: Stress-Aware Speech-to-Speech Translation with Emphasis Preservation

We propose a stress-aware speech-to-speech translation (S2ST) system that preserves word-level emphasis by leveraging LLMs for cross-lingual emphasis conversion. Our method translates source-language stress into target-language tags that guide a controllable TTS model. To overcome data scarcity, we developed a pipeline to automatically generate aligned training data and introduce the "LLM-as-Judge" for evaluation. Experiments show our approach substantially outperforms baselines in preserving emphasis while maintaining comparable translation quality, speaker intent, and naturalness. Our work highlights the importance of prosody in translation and provides an effective, data-efficient solution for preserving paralinguistic cues in S2ST.

cs.CL↗

arXiv2025

DPO-Tuned Large Language Models for Segmentation in Simultaneous Speech Translation

Simultaneous speech translation requires accurate segmentation to balance translation quality and latency. Recent studies such as SHAS have introduced pretrained segmentation models, achieving stronger performance than heuristic rules. However, segmentation models such as SHAS, though pretrained and more robust than heuristic methods, are still constrained by supervised learning objectives and do not incorporate human preference alignment, which is crucial for natural real-time interpretation. In this work, we propose a segmentation framework based on large language models (LLMs) trained with Direct Preference Optimization (DPO). By leveraging preference alignment, our method enables LLMs to predict natural segmentation points that better meet the demands of real-time translation. We evaluate the system on the ACL 60/60 corpus across three language pairs (English-Japanese, Chinese, German), using SeamlessM4T v2 as the translation backbone. Experimental results show that our DPO-tuned LLM achieves higher segmentation accuracy than SHAS and yields consistent improvements in translation quality (BLEU, COMET) as well as latency (Average Lagging). Furthermore, our system benefits from IWSLT baselines for direct comparison. These findings highlight the potential of preference-tuned LLMs to surpass existing pretrained segmentation models and advance adaptive, human-aligned simultaneous interpretation.

cs.CL↗

arXiv2025

TokenChain: A Discrete Speech Chain via Semantic Token Modeling

Machine Speech Chain, simulating the human perception-production loop, proves effective in jointly improving ASR and TTS. We propose TokenChain, a fully discrete speech chain coupling semantic-token ASR with a two-stage TTS: an autoregressive text-to-semantic model co-trained with ASR and a masked-generative semantic-to-acoustic model for synthesis only. End-to-end feedback across the text interface is enabled with straight-through argmax/Gumbel-Softmax and balanced with supervised ASR via dynamic weight averaging. Ablations examine optimal temperature schedules for in- and cross-domain transfer. Evaluation reveals TokenChain surpasses baseline accuracy 2-6 epochs earlier and yields 5-13% lower equal-epoch error with stable T2S on LibriSpeech, and reduces relative ASR WER by 56% and T2S WER by 31% on TED-LIUM with minimal forgetting, showing that chain learning remains effective with token interfaces and models.

eess.AS↗

arXiv2025

Redefining Machine Simultaneous Interpretation: From Incremental Translation to Human-Like Strategies

Simultaneous Machine Translation (SiMT) requires high-quality translations under strict real-time constraints, which traditional encoder-decoder policies with only READ/WRITE actions cannot fully address. We extend the action space of SiMT with four adaptive actions: SENTENCE_CUT, DROP, PARTIAL_SUMMARIZATION and PRONOMINALIZATION, which enable real-time restructuring, omission, and simplification while preserving semantic fidelity. We implement these actions in a decoder-only large language model (LLM) framework and construct training references through action-aware prompting. To evaluate both quality and latency, we further develop a latency-aware TTS pipeline that maps textual outputs to speech with realistic timing. Experiments on the ACL60/60 English-Chinese and English-German benchmarks show that our framework consistently improves semantic metrics (e.g., COMET-KIWI) and achieves lower delay (measured by Average Lagging) compared to reference translations and salami-based baselines. Notably, combining DROP and SENTENCE_CUT yields the best overall balance between fluency and latency. These results demonstrate that enriching the action space of LLM-based SiMT provides a promising direction for bridging the gap between human and machine interpretation.

cs.CL↗

arXiv2025

SASST: Leveraging Syntax-Aware Chunking and LLMs for Simultaneous Speech Translation

This work proposes a grammar-based chunking strategy that segments input streams into semantically complete units by parsing dependency relations (e.g., noun phrase boundaries, verb-object structures) and punctuation features. The method ensures chunk coherence and minimizes semantic fragmentation. Building on this mechanism, we present SASST (Syntax-Aware Simultaneous Speech Translation), an end-to-end framework integrating frozen Whisper encoder and decoder-only LLM. The unified architecture dynamically outputs translation tokens or symbols to jointly optimize translation timing and content, with target-side reordering addressing word-order divergence. Experiments on CoVoST2 multilingual corpus En-{De, Zh, Ja} demonstrate significant translation quality improvements across languages and validate the effectiveness of syntactic structures in LLM-driven SimulST systems.

cs.CL↗

arXiv2025

Modified extremal Kähler metrics and multiplier Hermitian-Einstein metrics

Motivated by the notion of multiplier Hermitian-Einstein metric of type $σ$ introduced by Mabuchi, we introduce the notion of $σ$-extremal Kähler metrics on compact Kähler manifolds, which generalizes Calabi's extremal Kähler metrics. We characterize the existence of this metric in terms of the coercivity of a certain functional on the space of Kähler metrics to show that, on a Fano manifold, the existence of a multiplier Hermitian-Einstein metric of type $σ$ implies the existence of a $σ$-extremal Kähler metric.

math.DG↗

arXiv2025

An Automatic Quality Metric for Evaluating Simultaneous Interpretation

Simultaneous interpretation (SI), the translation of one language to another in real time, starts translation before the original speech has finished. Its evaluation needs to consider both latency and quality. This trade-off is challenging especially for distant word order language pairs such as English and Japanese. To handle this word order gap, interpreters maintain the word order of the source language as much as possible to keep up with original language to minimize its latency while maintaining its quality, whereas in translation reordering happens to keep fluency in the target language. This means outputs synchronized with the source language are desirable based on the real SI situation, and it's a key for further progress in computational SI and simultaneous machine translation (SiMT). In this work, we propose an automatic evaluation metric for SI and SiMT focusing on word order synchronization. Our evaluation metric is based on rank correlation coefficients, leveraging cross-lingual pre-trained language models. Our experimental results on NAIST-SIC-Aligned and JNPC showed our metrics' effectiveness to measure word order synchronization between source and target language.

cs.CL↗