DHDaniel Hershcovich

Daniel Hershcovich

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Daniel Hershcovich has a total of 75 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Modeling Textual Emotions in Literary Fiction

    by Kirstine Nielsen Degn, Alexander Conroy, Xiaoyuan Jiang, Ali Al-Laith, Jens Bjerring-Hansen, Tanya Karoli Christensen, Ingo Zettler, Daniel Hershcovich

    This paper explores how finetuned transformer models can be used to examine the vast and underexplored emotional-historical dimensions of literary history. Drawing on a corpus of 859 nineteenth-century Danish and Norwegian novels, we develop an interdisciplinary framework for analyzing the expression of emotions in literary texts. This paper presents three main contributions: (1) we introduce an annotated multilabel dataset of emotion categories for 859 nineteenth-century Danish and Norwegian novels; (2) we evaluate the performance and generalization capabilities of large language models in classifying emotions in long and complex texts; and (3) we propose a theoretical framework that integrates insights from psychology, linguistics, literary theory, and NLP to support future research on cultural and historical formations of emotions. By combining computational methods with explicit humanistic theorization, this approach not only enables large-scale analysis of literature’s emotional dimensions but also demonstrates how methods used in the digital humanities can serve as a gateway to new theorization of emotions and their textual manifestations, offering valuable contributions to humanities scholars engaged in the so-called affective turn.

  • Open Access

    Cross-cultural value alignment frameworks for responsible AI governance: Evidence from China-West comparative analysis

    by Haijiang Liu, Jinguang Gu, Xun Wu, Daniel Hershcovich, Qiaoling Xiao

    As Large Language Models (LLMs) increasingly influence high-stakes decision-making across global contexts, ensuring their alignment with diverse cultural values has become a critical governance challenge. This study presents a Multi-Layered Auditing Platform for Responsible AI that systematically evaluates cross-cultural value alignment in China-origin and Western-origin LLMs through four integrated methodologies: Ethical Dilemma Corpus for assessing temporal stability, Diversity-Enhanced Framework (DEF) for quantifying cultural fidelity, First-Token Probability Alignment for distributional accuracy, and Multi-stAge Reasoning frameworK (MARK) for interpretable decision-making. Our comparative analysis of 20+ leading models, such as Qwen, GPT-4o, Claude, LLaMA, and DeepSeek, reveals universal challenges-fundamental instability in value systems, systematic under-representation of younger demographics, and non-linear relationships between model scale and alignment quality-alongside divergent regional development trajectories. While China-origin models increasingly emphasize multilingual data integration for context-specific optimization, Western models demonstrate greater architectural experimentation but persistent U.S.-centric biases. Neither paradigm achieves robust cross-cultural generalization. We establish that Mistral-series architectures significantly outperform LLaMA3-series in cross-cultural alignment, and that Full-Parameter Fine-Tuning on diverse datasets surpasses Reinforcement Learning from Human Feedback in preserving cultural variation...

  • Open Access

    HapticCap: A Multimodal Dataset and Task for Understanding User Experience of Vibration Haptic Signals

    by Guimin Hu, Daniel Hershcovich, Hasti Seifi

    Haptic signals, from smartphone vibrations to virtual reality touch feedback, can effectively convey information and enhance realism, but designing signals that resonate meaningfully with users is challenging. To facilitate this, we introduce a multimodal dataset and task, of matching user descriptions to vibration haptic signals, and highlight two primary challenges: (1) lack of large haptic vibration datasets annotated with textual descriptions as collecting haptic descriptions is time-consuming, and (2) limited capability of existing tasks and models to describe vibration signals in text. To advance this area, we create HapticCap, the first fully human-annotated haptic-captioned dataset, containing 92,070 haptic-text pairs for user descriptions of sensory, emotional, and associative attributes of vibrations. Based on HapticCap, we propose the haptic-caption retrieval task and present the results of this task from a supervised contrastive learning framework that brings together text representations within specific categories and vibrations. Overall, the combination of language model T5 and audio model AST yields the best performance in the haptic-caption retrieval task, especially when separately trained for each description category.


  • Open Access

    Unhappy Texts?

    by Kirstine Nielsen Degn, Jens Bjerring-Hansen, Ali Al-Laith, Daniel Hershcovich

    Our article discusses the hypothesis that the texts of women writers of the Modern Breakthrough in Scandinavia were particularly unhappy. We examine this common claim, along with some of the quantitative and qualitative issues it raises. Does this correlation of gender and affect hold true for the entire spectrum of women’s literary production from the era? What about male authorship and its affectivity? And what does ‘unhappy’ even mean? We confront this hypothesis and the associated questions through two interventions. The first is a quantification made possible by new digital archives and methodologies, which allow for a radical upscaling of the investigation’s empirical foundation. The second is to approach the nineteenth-century texts with a framework from the fields of gender studies and affect theory. Our findings are the following: (1) The thesis of the unhappy text appears partially true, but importantly, women are even more overrepresented among the positive texts. (2) The affect category of neutrality is more significant. Neutrality turns out to be a male, canonical enterprise, while low neutrality is primarily associated with forgotten or neglected women authors. The most crucial gender bias in the affective economy of the texts is the lack of neutrality in literature by women. (3) This and other biases point to clear intersectional dynamics between the author’s gender, the affective qualities and quantities of the texts, and their social status.


  • Open Access

    HapticCap: A Multimodal Dataset and Task for Understanding User Experience of Vibration Haptic Signals

    by Guimin Hu, Daniel Hershcovich, Hasti Seifi

    Guimin Hu, Daniel Hershcovich, Hasti Seifi. Findings of the Association for Computational Linguistics: EMNLP 2025. 2025.


  • Open Access

    Beyond Words: Exploring Cultural Value Sensitivity in Multimodal Models

    by Srishti Yadav, Zhi Zhang, Daniel Hershcovich, Ekaterina Shutova

    Srishti Yadav, Zhi Zhang, Daniel Hershcovich, Ekaterina Shutova. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.


  • Top Cited Papers
    Open Access

    Specializing Large Language Models to Simulate Survey Response Distributions for Global Populations

    by Yong Cao, Haijiang Liu, Arnav Arora, Isabelle Augenstein, Paul Röttger, Daniel Hershcovich

    Yong Cao, Haijiang Liu, Arnav Arora, Isabelle Augenstein, Paul Röttger, Daniel Hershcovich. Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers). 2025.


  • Open Access

    Bridging Cultural Nuances in Dialogue Agents through Cultural Value Surveys

    by Yong Cao, Min Chen, Daniel Hershcovich

    The cultural landscape of interactions with dialogue agents is a compelling yet relatively unexplored territory. It's clear that various sociocultural aspects -- from communication styles and beliefs to shared metaphors and knowledge -- profoundly impact these interactions. To delve deeper into this dynamic, we introduce cuDialog, a first-of-its-kind benchmark for dialogue generation with a cultural lens. We also develop baseline models capable of extracting cultural attributes from dialogue exchanges, with the goal of enhancing the predictive accuracy and quality of dialogue agents. To effectively co-learn cultural understanding and multi-turn dialogue predictions, we propose to incorporate cultural dimensions with dialogue encoding features. Our experimental findings highlight that incorporating cultural value surveys boosts alignment with references and cultural markers, demonstrating its considerable influence on personalization and dialogue quality. To facilitate further exploration in this exciting domain, we publish our benchmark publicly accessible at https://github.com/yongcaoplus/cuDialog.


  • Open Access

    Geo-Encoder: A Chunk-Argument Bi-Encoder Framework for Chinese Geographic Re-Ranking

    by Yong Cao, Ruixue Ding, Boli Chen, Xianzhi Li, Min Chen, Daniel Hershcovich, Pengjun Xie, Fei Huang

    Chinese geographic re-ranking task aims to find the most relevant addresses among retrieved candidates, which is crucial for location-related services such as navigation maps. Unlike the general sentences, geographic contexts are closely intertwined with geographical concepts, from general spans (e.g., province) to specific spans (e.g., road). Given this feature, we propose an innovative framework, namely Geo-Encoder, to more effectively integrate Chinese geographical semantics into re-ranking pipelines. Our methodology begins by employing off-the-shelf tools to associate text with geographical spans, treating them as chunking units. Then, we present a multi-task learning module to simultaneously acquire an effective attention matrix that determines chunk contributions to extra semantic representations. Furthermore, we put forth an asynchronous update mechanism for the proposed addition task, aiming to guide the model capable of effectively focusing on specific chunks. Experiments on two distinct Chinese geographic re-ranking datasets, show that the Geo-Encoder achieves significant improvements when compared to state-of-the-art baselines. Notably, it leads to a substantial improvement in the Hit@1 score of MGEO-BERT, increasing it by 6.22% from 62.76 to 68.98 on the GeoTES dataset.