arXiv subjects

Christian Wolff

Publications and source records attributed to Christian Wolff.

At least 19 recordsLinked to original sources

arXiv2026

Prompting Is All You Need: Multi-view Prompting Large Language Models for Aspect-Based Sentiment Analysis

Recent work explored the capabilities of Large Language Models (LLMs) in Aspect-Based Sentiment Analysis (ABSA) through few-shot prompting, requiring substantially fewer annotated examples while achieving notable improvements over zero-shot baselines. However, a performance gap remained compared to models fine-tuned on hundreds of examples, and the computational costs of LLM inference present practical barriers to deployment. We introduce LLM-based Multi-View Prompting (LLM-MvP), which adapts the multi-view principle of considering multiple element orderings to LLM prompting. By combining schema-constrained decoding with a context-free grammar and prefix batching, LLM-MvP achieves performance competitive or superior to fine-tuned approaches while substantially reducing computational overhead. Extensive experiments across five benchmark datasets demonstrate that LLM-MvP closes the gap between few-shot prompting and fine-tuned models, offering a practical and efficient solution for ABSA.

cs.CL↗

arXiv2026

Annotation Quality in Aspect-Based Sentiment Analysis: A Case Study Comparing Experts, Students, Crowdworkers, and Large Language Model

Aspect-Based Sentiment Analysis (ABSA) enables fine-grained opinion analysis by identifying sentiments toward specific aspects or targets within a text. While ABSA has been widely studied for English, research on other languages such as German remains limited, largely due to the lack of high-quality annotated datasets. This paper examines how different annotation sources influence the development of German ABSA. To this end, an existing dataset is re-annotated by experts to establish a ground truth, which serves as a reference for evaluating annotations produced by students, crowdworkers, Large Language Models (LLMs), and experts. Annotation quality is compared using Inter-Annotator Agreement (IAA) and its impact on downstream model performance for different ABSA subtasks. The evaluation focuses on Aspect Category Sentiment Analysis (ACSA) and Target Aspect Sentiment Detection (TASD). We apply State-of-the-Art (SOTA) methods for ABSA, including BERT-, T5-, and LLaMA-based approaches to assess performance differences, spanning fine-tuning and in-context learning with instruction prompts. The findings provide practical insights into trade-offs between annotation reliability and efficiency, offering guidance for dataset construction in under-resourced Natural Language Processing (NLP) scenarios.

cs.CL↗

arXiv2026

Zero-Shot to Full-Resource: Cross-lingual Transfer Strategies for Aspect-Based Sentiment Analysis

Aspect-based Sentiment Analysis (ABSA) extracts fine-grained opinions toward specific aspects within text but remains largely English-focused despite major advances in transformer-based and instruction-tuned models. This work presents a multilingual evaluation of state-of-the-art ABSA approaches across seven languages (English, German, French, Dutch, Russian, Spanish, and Czech) and four subtasks (ACD, ACSA, TASD, ASQP). We systematically compare different transformer architectures under zero-resource, data-only, and full-resource settings, using cross-lingual transfer, code-switching and machine translation. Fine-tuned Large Language Models (LLMs) achieve the highest overall scores, particularly in complex generative tasks, while few-shot counterparts approach this performance in simpler setups, where smaller encoder models also remain competitive. Cross-lingual training on multiple non-target languages yields the strongest transfer for fine-tuned LLMs, while smaller encoder or seq-to-seq models benefit most from code-switching, highlighting architecture-specific strategies for multilingual ABSA. We further contribute two new German datasets, an adapted GERestaurant and the first German ASQP dataset (GERest), to encourage multilingual ABSA research beyond English.

cs.CL↗

arXiv2026

Seeing Candidates at Scale: Multimodal LLMs for Visual Political Communication on Instagram

This paper presents a computational case study that evaluates the capabilities of specialized machine learning models and emerging multimodal large language models for Visual Political Communication (VPC) analysis. Focusing on concentrated visibility in Instagram stories and posts during the 2021 German federal election campaign, we compare the performance of traditional computer vision models (FaceNet512, RetinaFace, Google Cloud Vision) with a multimodal large language model (GPT-4o) in identifying front-runner politicians and counting individuals in images. GPT-4o outperformed the other models, achieving a macro F1-score of 0.89 for face recognition and 0.86 for person counting in stories. These findings demonstrate the potential of advanced AI systems to scale and refine visual content analysis in political communication while highlighting methodological considerations for future research.

cs.CV↗

arXiv2026

AnnoABSA: A Web-Based Annotation Tool for Aspect-Based Sentiment Analysis with Retrieval-Augmented Suggestions

We introduce AnnoABSA, the first web-based annotation tool to support the full spectrum of Aspect-Based Sentiment Analysis (ABSA) tasks. The tool is highly customizable, enabling flexible configuration of sentiment elements and task-specific requirements. Alongside manual annotation, AnnoABSA provides optional Large Language Model (LLM)-based retrieval-augmented generation (RAG) suggestions that offer context-aware assistance in a human-in-the-loop approach, keeping the human annotator in control. To improve prediction quality over time, the system retrieves the ten most similar examples that are already annotated and adds them as few-shot examples in the prompt, ensuring that suggestions become increasingly accurate as the annotation process progresses. Released as open-source software under the MIT License, AnnoABSA is freely accessible and easily extendable for research and practical applications.

cs.CL↗

arXiv2026

LLM-as-an-Annotator: Training Lightweight Models with LLM-Annotated Examples for Aspect Sentiment Tuple Prediction

Training models for Aspect-Based Sentiment Analysis (ABSA) tasks requires manually annotated data, which is expensive and time-consuming to obtain. This paper introduces LA-ABSA, a novel approach that leverages Large Language Model (LLM)-generated annotations to fine-tune lightweight models for complex ABSA tasks. We evaluate our approach on five datasets for Target Aspect Sentiment Detection (TASD) and Aspect Sentiment Quad Prediction (ASQP). Our approach outperformed previously reported augmentation strategies and achieved competitive performance with LLM-prompting in low-resource scenarios, while providing substantial energy efficiency benefits. For example, using 50 annotated examples for in-context learning (ICL) to guide the annotation of unlabeled data, LA-ABSA achieved an F1 score of 49.85 for ASQP on the SemEval Rest16 dataset, closely matching the performance of ICL prompting with Gemma-3-27B (51.10), while requiring significantly lower computational resources.

cs.CL↗

arXiv2026

nchellwig at SemEval-2026 Task 3: Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis using Large Language Models

We present Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis in SemEval-2026 Task 3 (Track A). SCSG enhances prediction reliability by executing a LoRA-adapted large language model multiple times per instance, retaining only tuples that achieve a majority consensus across runs. To mitigate the computational overhead of multiple forward passes, we leverage vLLM's PagedAttention mechanism for efficient key--value cache reuse. Evaluation across 6 languages and 8 language--domain combinations demonstrates that self-consistency with 15 executions yields statistically significant improvements over single-inference prompting, with our system (leveraging Gemma 3) ranking in the top seven across all settings, achieving second place on three out of four English subsets and first place on Tatar-Restaurant for DimASTE.

cs.CL↗

arXiv2025

Going off Pattern? QAOA Parameter Heuristics and Potentials of Parsimony

Structured variational quantum algorithms such as the Quantum Approximate Optimisation Algorithm (QAOA) have emerged as leading candidates for exploiting advantages of near-term quantum hardware. They interlace classical computation, in particular optimisation of variational parameters, with quantum-specific routines, and combine problem-specific advantages -- sometimes even provable -- with adaptability to the constraints of noisy, intermediate-scale quantum (NISQ) devices. While circuit depth can be parametrically increased and is known to improve performance in an ideal (noiseless) setting, on realistic hardware greater depth exacerbates noise: The overall quality of results depends critically on both, variational parameters and circuit depth. Although identifying optimal parameters is NP-hard, prior work has suggested that they may exhibit regular, predictable patterns for increasingly deep circuits and depending on the studied class of problems. In this work, we systematically investigate the role of classical parameters in QAOA performance through extensive numerical simulations and suggest a simple, yet effective heuristic scheme to find good parameters for low-depth circuits. Our results demonstrate that: (i) high-quality parameters often deviate substantially from expected patterns; (ii) QAOA performance becomes progressively less sensitive to specific parameter choices as depth increases; and (iii) iterative component-wise fixing performs on par with, and at shallow depth may even outperform, several established parameter-selection strategies. We identify conditions under which structured parameter patterns emerge, and when deviations from the patterns warrant further consideration. These insights for low-depth circuits may inform more robust pathways to harnessing QAOA in realistic quantum computing scenarios.

quant-ph↗

arXiv2025

Do we still need Human Annotators? Prompting Large Language Models for Aspect Sentiment Quad Prediction

Aspect sentiment quad prediction (ASQP) facilitates a detailed understanding of opinions expressed in a text by identifying the opinion term, aspect term, aspect category and sentiment polarity for each opinion. However, annotating a full set of training examples to fine-tune models for ASQP is a resource-intensive process. In this study, we explore the capabilities of large language models (LLMs) for zero- and few-shot learning on the ASQP task across five diverse datasets. We report F1 scores almost up to par with those obtained with state-of-the-art fine-tuned models and exceeding previously reported zero- and few-shot performance. In the 20-shot setting on the Rest16 restaurant domain dataset, LLMs achieved an F1 score of 51.54, compared to 60.39 by the best-performing fine-tuned method MVP. Additionally, we report the performance of LLMs in target aspect sentiment detection (TASD), where the F1 scores were close to fine-tuned models, achieving 68.93 on Rest16 in the 30-shot setting, compared to 72.76 with MVP. While human annotators remain essential for achieving optimal performance, LLMs can reduce the need for extensive manual annotation in ASQP tasks.

cs.CL↗

arXiv2024

Detecting Calls to Action in Multimodal Content: Analysis of the 2021 German Federal Election Campaign on Instagram

This study investigates the automated classification of Calls to Action (CTAs) within the 2021 German Instagram election campaign to advance the understanding of mobilization in social media contexts. We analyzed over 2,208 Instagram stories and 712 posts using fine-tuned BERT models and OpenAI's GPT-4 models. The fine-tuned BERT model incorporating synthetic training data achieved a macro F1 score of 0.93, demonstrating a robust classification performance. Our analysis revealed that 49.58% of Instagram posts and 10.64% of stories contained CTAs, highlighting significant differences in mobilization strategies between these content types. Additionally, we found that FDP and the Greens had the highest prevalence of CTAs in posts, whereas CDU and CSU led in story CTAs.

cs.SI↗

arXiv2024

Preserving the Ephemeral: Instagram Story Archiving with the Tidal Tales Plugin

We introduce the Tidal Tales Plugin, a Firefox extension for efficiently collecting and archiving of Instagram stories, addressing the challenges of ephemeral data in social media research. It enables an automated collection of story metadata and media files without risking account bans. It contributes to Web Science by facilitating expansive, long-term studies with enhanced data access and integrity.

cs.SI↗

arXiv2024

GERestaurant: A German Dataset of Annotated Restaurant Reviews for Aspect-Based Sentiment Analysis

We present GERestaurant, a novel dataset consisting of 3,078 German language restaurant reviews manually annotated for Aspect-Based Sentiment Analysis (ABSA). All reviews were collected from Tripadvisor, covering a diverse selection of restaurants, including regional and international cuisine with various culinary styles. The annotations encompass both implicit and explicit aspects, including all aspect terms, their corresponding aspect categories, and the sentiments expressed towards them. Furthermore, we provide baseline scores for the four ABSA tasks Aspect Category Detection, Aspect Category Sentiment Analysis, End-to-End ABSA and Target Aspect Sentiment Detection as a reference point for future advances. The dataset fills a gap in German language resources and facilitates exploration of ABSA in the restaurant domain.

cs.CL↗

arXiv2024

Innovations in Cover Song Detection: A Lyrics-Based Approach

Cover songs are alternate versions of a song by a different artist. Long being a vital part of the music industry, cover songs significantly influence music culture and are commonly heard in public venues. The rise of online music platforms has further increased their prevalence, often as background music or video soundtracks. While current automatic identification methods serve adequately for original songs, they are less effective with cover songs, primarily because cover versions often significantly deviate from the original compositions. In this paper, we propose a novel method for cover song detection that utilizes the lyrics of a song. We introduce a new dataset for cover songs and their corresponding originals. The dataset contains 5078 cover songs and 2828 original songs. In contrast to other cover song datasets, it contains the annotated lyrics for the original song and the cover song. We evaluate our method on this dataset and compare it with multiple baseline approaches. Our results show that our method outperforms the baseline approaches.

cs.IR↗

arXiv2023

Advancing High-Throughput Combinatorial Aging Studies of Hybrid Perovskite Thin-Films via Precise Automated Characterization Methods and Machine Learning Assisted Analysis

To optimize materials' stability, automated high-throughput workflows are of increasing interest. However, many of those workflows use processes not suitable for large-area depositions which limits the transferability of results. While combinatorial approaches based on vapour-based depositions are inherently scalable, their potential for controlled stability assessments has yet to be exploited. Based on MAPbI3 thin-films as a prototypical system, we demonstrate a combinatorial inert-gas workflow to study materials degradation based on intrinsic factors only, closely resembling conditions in encapsulated de-vices. Through a comprehensive set of automated X-Ray fluorescence (XRF), X-Ray diffraction (XRD) and UV-Vis characterizations, we aim to obtain a holistic understanding of thin-film properties of pristine and aged thin-films. From phase changes derived from XRD characterizations before and after aging, we observe simi-lar aging behaviours for MAPbI3 thin-films with varying PbI2 residuals. Using a custom-designed in-situ UV-Vis aging setup, the combinatorial libraries are exposed to relevant aging conditions, such as heat or light-bias exposure. Simultaneously, UV-Vis photospectroscopy is performed to gain kinetic insights into the aging process which can be linked to intrinsic degradation processes such as autocatalytic decomposition. Despite scattering effects, which complicate the conventional interpretation of in-situ UV-Vis results, we demonstrate how a machine learning model trained on the comprehensive characterization data before and after the aging process can link optical changes to phase changes during aging. Consequently, this approach does not only enable semi-quantitative comparisons of materials' stability but also provides detailed insights into the underlying degradation processes which are otherwise mostly reported for investigations on single samples.

physics.app-ph↗

arXiv2023

Self-hybridisation between interband transitions and Mie modes in dielectric nanoparticles

We discuss the possibility of self-hybridisation in high-index dielectric nanoparticles, where Mie modes of electric or magnetic type can couple to the interband transitions of the material, leading to spectral anticrossings. Starting with an idealised system described by moderately high constant permittivity with a narrow Lorentzian, in which self-hybridisation is visible for both plane-wave and electron-beam excitation, we embark on a quest for realistic systems where this effect should be visible. We explore a variety of spherical particles made of traditional semiconductors such as Si, GaAs, and GaP. With the effect hardly discernible, we identify two major causes hindering observation of self-hybridisation: the very broad spectral fingerprints of interband transitions in most candidate materials, and the significant overlap between electric and magnetic Mie modes in nanospheres. We thus depart from the spherical shape, and show that interband--Mie hybridisation is indeed feasible in the example of GaAs cylinders, even with a simple plane-wave source. This so-far unreported kind of polariton has to be considered when interpreting experimental spectra of Mie-resonant nanoparticles and assigning modal characters to specific features. On the other hand, it has the potential to be useful for the characterisation of the optical properties of dielectric materials, through control of the hybridisation strength via nanoparticle size and shape.

cond-mat.mes-hall↗

arXiv2023

Electron beams traversing spherical nanoparticles: analytic and numerical treatment

We present an analytic, Mie theory-based solution for the energy-loss and the photon-emission probabilities in the interaction of spherical nanoparticles with electrons passing nearby and through them, in both cathodoluminescence and electron energy-loss spectroscopies. In particular, we focus on the case of penetrating electron trajectories, for which the complete fully electrodynamic and relativistic formalism has not been reported as yet. We exhibit the efficiency of this method in describing collective excitations in matter through calculations for a dispersive and lossy system, namely a sphere described by a Drude permittivity. Subsequently, we use the analytic solution to corroborate the implementation of electron-beam sources in a state-of-the-art numerical method for problems in electrodynamics, the discontinuous Galerkin time-domain (DGTD) method. We show that the two approaches produce spectra in good mutual agreement, and demonstrate the versatility of DGTD via simulations of spherical nanoparticles characterized by surface roughness. The possibility of simultaneously employing both kinds of calculations (analytic and numerical) facilitates a better understanding of the rich optical response of nanophotonic architectures excited by fast electron beams.

cond-mat.mes-hall↗

arXiv2023

Inducing room-temperature valley polarization of excitonic emission in transition metal dichalcogenide monolayers

The lowest energy states in transition metal dichalcogenide (TMD) monolayers follow valley selection rules, which have attracted vast interest due to the possibility of encoding and processing of quantum information. However, these quantum states are strongly affected by the temperature-dependent intervalley scattering causing complete valley depolarization, which is hampering any practical applications of TMD monolayers at room temperature. Therefore, for achieving clear and robust valley polarization in TMD monolayers one needs to suppress parasitic depolarization processes, which is the central challenge in the growing field of valleytronics. Here, in electron-doping experiments on TMD monolayers, we demonstrate that strong doping levels beyond $10^{13}$~cm$^{-2}$ can induce 61\% and 37\% valley contrast at room temperature in tungsten diselenide and molybdenum diselenide monolayers, respectively. Our results indicate that charged excitons in TMD monolayers can be utilized as quantum units in designing of practical valleytronic devices operating at 300 K.

cond-mat.mes-hall↗

arXiv2023

Direct electrical modulation of surface response in a single plasmonic nanoresonator

Classical electrodynamics describes the optical response of systems using bulk electronic properties and infinitesimally thin boundaries. However, due to the quantum nature of electrons, interfaces have a finite thickness. Non-classical surface effects become increasingly important as ever smaller nanoscale systems are realized and eventually dominate over volume-related phenomena. Investigating the response of surface electrons in such systems, therefore, becomes imperative. One way to gain control over non-classical interface effects and study them is through electrical gating, as the static screening charges reside exclusively at the surface. Here, we investigate the modulation of the surface response upon direct electric charging of a single plasmonic nanoresonator by measuring the resulting changes in resonance. We analyze the observed effects within the general framework of surface-response functions and provide a basic model derived from electron spill-out within the local-response approximation (LRA). Our observed change in resonance frequency is well accounted for by assuming a modulation of the in-plane surface current. Surprisingly, we also measure a change in the resonance width, where adding electrons to the surface leads to a narrowing of the plasmonic resonance, i.e., reduced losses. The description of such effects requires considering nonlocal effects and the inclusion of a possible anisotropy of the perturbed surface permittivity. Our experiment, therefore, opens a vast field of investigations on how to gain control over the surface response in plasmonic resonators and to develop ultrafast and extremely small electrically driven plasmonic modulators and metasurfaces by leveraging electrical control over non-classical surface effects.

physics.optics↗