Selected work

Representative Papers

Robustness of Graph Self-Supervised Learning to Real-World Noise: A Case Study on Text-Driven Biomedical Graphs

May 06, 2026

This study addresses the vulnerability of graph self-supervised learning (GSSL) to noise inherent in real-world, text-derived biomedical knowledge graphs—a challenge overlooked by existing research that predominantly evaluates methods on clean or synthetic graphs. To bridge this gap, the authors introduce the first benchmarking framework tailored to realistic noisy settings, systematically comparing model performance on MedMentions (a noisy graph) against UMLS (a curated, clean graph). Through comprehensive analysis of pretraining tasks and GNN architectures, they find that feature reconstruction exhibits robustness under noise and that bidirectional message passing consistently outperforms unidirectional variants. The work further proposes NATD-GSSL, a unified pipeline integrating graph construction, refinement, and representation learning, which achieves up to a 7% improvement over language model baselines. Code and benchmark datasets are publicly released.

0 citationsRead paper

A Unified Benchmark for Evaluating Knowledge Graph Construction Methods and Graph Neural Networks

May 06, 2026

Existing automatically constructed knowledge graphs often suffer from noise, fragmentation, and semantic inconsistencies, making it difficult to disentangle whether performance differences in graph neural networks stem from the models themselves or from variations in graph quality. To address this, this work proposes the first dual-objective benchmark that generates multiple automatically constructed graphs from the same biomedical text corpus and incorporates an expert-annotated high-quality reference graph. Through semi-supervised node classification tasks, the framework jointly evaluates the effectiveness of knowledge graph construction methods and the robustness of graph neural networks under realistic noise conditions. This approach enables standardized, reproducible, and scalable co-evaluation, facilitating fair comparisons across graph construction techniques and revealing the upper performance bounds of downstream models.

0 citationsRead paper

A Paradigm for Interpreting Metrics and Identifying Critical Errors in Automatic Speech Recognition

May 05, 2026

Traditional automatic speech recognition evaluation metrics, such as word error rate (WER) and character error rate (CER), fail to capture human perception of errors and neglect linguistic and semantic influences. This work proposes a novel paradigm that embeds any perception-oriented evaluation metric into the minimum edit distance (minED) framework to produce an intuitively interpretable equivalent error rate. For the first time, this approach translates human perceptual modeling into a comprehensible error rate format, enabling quantification of error severity from the perspective of human understanding. The resulting metric not only aligns closely with human judgments but also effectively identifies recognition errors that critically impact semantic comprehension.

0 citationsRead paper

A Comprehensive Analysis of Tokenization and Self-Supervised Learning in End-to-End Automatic Speech Recognition applied on French Language

May 05, 2026

This study addresses the limitations of conventional character- or word-error-rate metrics in evaluating end-to-end automatic speech recognition (ASR) systems, which often fail to capture the full spectrum of transcription quality. Focusing on French, the work proposes a multidimensional evaluation framework that integrates both linguistic and acoustic perspectives to overcome the constraints of single-metric assessments. Through systematic comparisons of various subword tokenization strategies—such as Byte Pair Encoding (BPE)—and prominent self-supervised speech representation models within end-to-end ASR architectures, the research elucidates how these components influence transcription accuracy and fluency. The resulting framework not only offers a more comprehensive and application-oriented approach to ASR evaluation but also establishes an empirical foundation for optimizing downstream French ASR systems.

0 citationsRead paper

Qualitative Evaluation of Language Model Rescoring in Automatic Speech Recognition

Apr 30, 2026

This study addresses the limitations of traditional automatic speech recognition (ASR) evaluation, which relies heavily on word error rate (WER) and fails to capture the grammatical and semantic characteristics of transcription errors. To overcome this, the authors propose two novel metrics: Part-of-Speech Error Rate (POSER) and Embedding Error Rate (EmbER), which quantify ASR output quality from the perspectives of grammatical correctness and semantic fidelity, respectively. By integrating language model rescoring, part-of-speech tagging, and semantic distance computation based on word embeddings, they construct a multidimensional qualitative evaluation framework. Experimental results demonstrate that these new metrics effectively reveal the contribution of language models to improving linguistic quality in transcriptions, thereby compensating for WER’s insufficiency in linguistic analysis.

0 citationsRead paper

Recent publications

Latest Papers

Robustness of Graph Self-Supervised Learning to Real-World Noise: A Case Study on Text-Driven Biomedical Graphs

May 06, 2026

This study addresses the vulnerability of graph self-supervised learning (GSSL) to noise inherent in real-world, text-derived biomedical knowledge graphs—a challenge overlooked by existing research that predominantly evaluates methods on clean or synthetic graphs. To bridge this gap, the authors introduce the first benchmarking framework tailored to realistic noisy settings, systematically comparing model performance on MedMentions (a noisy graph) against UMLS (a curated, clean graph). Through comprehensive analysis of pretraining tasks and GNN architectures, they find that feature reconstruction exhibits robustness under noise and that bidirectional message passing consistently outperforms unidirectional variants. The work further proposes NATD-GSSL, a unified pipeline integrating graph construction, refinement, and representation learning, which achieves up to a 7% improvement over language model baselines. Code and benchmark datasets are publicly released.

0 citationsRead paper

A Unified Benchmark for Evaluating Knowledge Graph Construction Methods and Graph Neural Networks

May 06, 2026

Existing automatically constructed knowledge graphs often suffer from noise, fragmentation, and semantic inconsistencies, making it difficult to disentangle whether performance differences in graph neural networks stem from the models themselves or from variations in graph quality. To address this, this work proposes the first dual-objective benchmark that generates multiple automatically constructed graphs from the same biomedical text corpus and incorporates an expert-annotated high-quality reference graph. Through semi-supervised node classification tasks, the framework jointly evaluates the effectiveness of knowledge graph construction methods and the robustness of graph neural networks under realistic noise conditions. This approach enables standardized, reproducible, and scalable co-evaluation, facilitating fair comparisons across graph construction techniques and revealing the upper performance bounds of downstream models.

0 citationsRead paper

A Paradigm for Interpreting Metrics and Identifying Critical Errors in Automatic Speech Recognition

May 05, 2026

Traditional automatic speech recognition evaluation metrics, such as word error rate (WER) and character error rate (CER), fail to capture human perception of errors and neglect linguistic and semantic influences. This work proposes a novel paradigm that embeds any perception-oriented evaluation metric into the minimum edit distance (minED) framework to produce an intuitively interpretable equivalent error rate. For the first time, this approach translates human perceptual modeling into a comprehensible error rate format, enabling quantification of error severity from the perspective of human understanding. The resulting metric not only aligns closely with human judgments but also effectively identifies recognition errors that critically impact semantic comprehension.

0 citationsRead paper

A Comprehensive Analysis of Tokenization and Self-Supervised Learning in End-to-End Automatic Speech Recognition applied on French Language

May 05, 2026

This study addresses the limitations of conventional character- or word-error-rate metrics in evaluating end-to-end automatic speech recognition (ASR) systems, which often fail to capture the full spectrum of transcription quality. Focusing on French, the work proposes a multidimensional evaluation framework that integrates both linguistic and acoustic perspectives to overcome the constraints of single-metric assessments. Through systematic comparisons of various subword tokenization strategies—such as Byte Pair Encoding (BPE)—and prominent self-supervised speech representation models within end-to-end ASR architectures, the research elucidates how these components influence transcription accuracy and fluency. The resulting framework not only offers a more comprehensive and application-oriented approach to ASR evaluation but also establishes an empirical foundation for optimizing downstream French ASR systems.

0 citationsRead paper

Qualitative Evaluation of Language Model Rescoring in Automatic Speech Recognition

Apr 30, 2026

This study addresses the limitations of traditional automatic speech recognition (ASR) evaluation, which relies heavily on word error rate (WER) and fails to capture the grammatical and semantic characteristics of transcription errors. To overcome this, the authors propose two novel metrics: Part-of-Speech Error Rate (POSER) and Embedding Error Rate (EmbER), which quantify ASR output quality from the perspectives of grammatical correctness and semantic fidelity, respectively. By integrating language model rescoring, part-of-speech tagging, and semantic distance computation based on word embeddings, they construct a multidimensional qualitative evaluation framework. Experimental results demonstrate that these new metrics effectively reveal the contribution of language models to improving linguistic quality in transcriptions, thereby compensating for WER’s insufficiency in linguistic analysis.

0 citationsRead paper