Multimodal affective computing underpins key tasks such as sentiment analysis and emotion recognition. Standard evaluations, however, often assume that textual, acoustic, and visual modalities are equally available. In real applications, some modalities are systematically more fragile or expensive, creating imbalanced missing rates and training biases that task-level metrics alone do not reveal. We introduce MissBench, a benchmark and framework for multimodal affective tasks that standardizes both shared and imbalanced missing-rate protocols on four widely used sentiment and emotion datasets. MissBench also defines two diagnostic metrics. The Modality Equity Index (MEI) measures how fairly different modalities contribute across missing-modality configurations. The Modality Learning Index (MLI) quantifies optimization imbalance by comparing modality-specific gradient norms during training, aggregated across modality-related modules. Experiments on representative method families show that models that appear robust under shared missing rates can still exhibit marked modality inequity and optimization imbalance under imbalanced conditions. These findings position MissBench, together with MEI and MLI, as practical tools for stress-testing and analyzing multimodal affective models in realistic incomplete-modality this http URL reproducibility, we release our code at: this https URL
多模态情感计算支撑着诸如情感分析和情绪识别等关键任务。然而,标准评估通常假设文本、音频和视觉模式是同样可用的。在实际应用中,某些模式由于脆弱性或成本原因会系统性地缺失,导致不平衡的丢失率和训练偏差,这些单靠任务级指标无法揭示。 我们推出了MissBench,这是一个针对多模态情感任务的标准基准和框架,涵盖了四种常用的情感和情绪数据集上的共享及不平衡丢失率协议。MissBench还定义了两个诊断指标:模式公平指数(Modality Equity Index, MEI)衡量不同模式在缺失模式配置下贡献的公正性;模式学习指数(Modality Learning Index, MLI)通过比较训练期间各模态特定的梯度范数来量化优化不平衡,这些梯度是跨模态相关模块聚合得到的。 代表性的方法族实验表明,在共享丢失率条件下看似稳健的模型在不平衡条件下面临显著的模式不公平性和优化不平衡。这一发现使MissBench及其MEI和MLI成为测试多模态情感模型在实际不完整模式情况下的压力工具及分析工具,以提高研究工作的可重复性。 为了确保再现性,我们发布了我们的代码:[这里可以填写发布地址]
https://arxiv.org/abs/2603.09874
Multimodal sentiment analysis aims to understand human emotions by integrating textual, auditory, and visual modalities. Although Multimodal Large Language Models (MLLMs) have achieved state-of-the-art performance via supervised fine-tuning (SFT), their end-to-end "black-box" nature limits interpretability. Existing methods incorporating Chain-of-Thought (CoT) reasoning are hindered by high annotation costs, while Reinforcement Learning (RL) faces challenges such as low exploration efficiency and sparse rewards, particularly on hard samples. To address these issues, we propose a novel training framework that integrates structured Discrimination-Calibration (DC) reasoning with Hint-based Reinforcement Learning. First, we perform cold-start SFT using high-quality CoT data synthesized by a teacher model (Qwen3Omni-30B), which inherently contains the DC structure. This equips the model with a reasoning paradigm that performs macro discrimination followed by fine-grained calibration from the initial stage. Building on this, we propose Hint-GRPO, which leverages the discrimination phase within the DC structure as a verifiable anchor during RL to provide directional hints for hard samples, guiding policy optimization and effectively mitigating the reward sparsity problem. Experiments on the Qwen2.5Omni-7B model demonstrate that our method not only achieves higher accuracy in fine-grained sentiment regression tasks but also generates high-quality structured reasoning chains. Crucially, it exhibits superior generalization capability in cross-domain evaluations. This enhances model interpretability while validating the positive contribution of explicit reasoning steps to model robustness, offering a new paradigm for building trustworthy and efficient sentiment analysis systems.
多模态情感分析旨在通过整合文本、听觉和视觉模态来理解人类情感。尽管多模态大语言模型(MLLMs)已通过监督微调(SFT)达到最先进性能,但其端到端的“黑箱”特性限制了可解释性。现有结合思维链(CoT)推理的方法受限于高昂的标注成本,而强化学习(RL)则面临探索效率低、奖励稀疏等挑战,尤其在困难样本上表现明显。针对这些问题,我们提出一种新颖的训练框架,将结构化的判别-校准(DC)推理与基于提示的强化学习相结合。首先,我们使用教师模型(Qwen3Omni-30B)合成的高质量CoT数据进行冷启动SFT,这些数据天然包含DC结构,使模型从初始阶段就具备“宏观判别+细粒度校准”的推理范式。在此基础上,我们提出Hint-GRPO方法,利用DC结构中的判别阶段作为RL中的可验证锚点,为困难样本提供方向性提示,引导策略优化,有效缓解奖励稀疏问题。在Qwen2.5Omni-7B模型上的实验表明,我们的方法不仅在细粒度情感回归任务中取得更高准确率,还能生成高质量的结构化推理链,并在跨域评估中展现出卓越的泛化能力。这既提升了模型可解释性,也验证了显式推理步骤对模型鲁棒性的积极贡献,为构建可信高效的情感分析系统提供了新范式。
https://arxiv.org/abs/2604.00013
Large language models are routinely deployed on text that varies widely in emotional tone, yet their reasoning behavior is typically evaluated without accounting for emotion as a source of representational variation. Prior work has largely treated emotion as a prediction target, for example in sentiment analysis or emotion classification. In contrast, we study emotion as a latent factor that shapes how models attend to and reason over text. We analyze how emotional tone systematically alters attention geometry in transformer models, showing that metrics such as locality, center-of-mass distance, and entropy vary across emotions and correlate with downstream question-answering performance. To facilitate controlled study of these effects, we introduce Affect-Uniform ReAding QA (AURA-QA), a question-answering dataset with emotionally balanced, human-authored context passages. Finally, an emotional regularization framework is proposed that constrains emotion-conditioned representational drift during training. Experiments across multiple QA benchmarks demonstrate that this approach improves reading comprehension in both emotionally-varying and non-emotionally varying datasets, yielding consistent gains under distribution shift and in-domain improvements on several benchmarks.
大型语言模型通常部署在情感基调迥异的文本上,但它们的推理行为一般不考虑情感作为表示变化的一个来源。以往的研究大多将情感视为预测目标,例如在情感分析或情绪分类中。相比之下,我们研究情感作为一种潜在因素,它塑造了模型如何关注和处理文本。我们分析了情感基调系统地改变变压器模型注意力几何的方式,并表明诸如局部性、质心距离和熵等度量标准会因情绪不同而变化,并且与下游问答任务的表现相关联。 为了便于这些效应的受控研究,我们引入了Affect-Uniform ReAding QA(AURA-QA),这是一个包含情感平衡的人工创作背景文本的问答数据集。最后,我们提出了一个情感正则化框架,在训练过程中限制由情感条件触发的表示漂移。在多个问答基准上的实验表明,这种方法提高了情感变化和非情感变化数据集中阅读理解的表现,并且在分布转变下以及几个基准内部都带来了持续的进步。 总的来说,这项研究揭示了大型语言模型如何受制于文本的情感因素,并提出了一种新的方法来改善这些模型的鲁棒性和准确性。
https://arxiv.org/abs/2603.09205
Multimodal Sentiment Analysis (MSA) seeks to infer human emotions by integrating textual, acoustic, and visual cues. However, existing approaches often rely on all modalities are completeness, whereas real-world applications frequently encounter noise, hardware failures, or privacy restrictions that result in missing modalities. There exists a significant feature misalignment between incomplete and complete modalities, and directly fusing them may even distort the well-learned representations of the intact modalities. To this end, we propose PRLF, a Progressive Representation Learning Framework designed for MSA under uncertain missing-modality conditions. PRLF introduces an Adaptive Modality Reliability Estimator (AMRE), which dynamically quantifies the reliability of each modality using recognition confidence and Fisher information to determine the dominant modality. In addition, the Progressive Interaction (ProgInteract) module iteratively aligns the other modalities with the dominant one, thereby enhancing cross-modal consistency while suppressing noise. Extensive experiments on CMU-MOSI, CMU-MOSEI, and SIMS verify that PRLF outperforms state-of-the-art methods across both inter- and intra-modality missing scenarios, demonstrating its robustness and generalization capability.
多模态情感分析(MSA)旨在通过整合文本、音频和视觉线索来推断人类情绪。然而,现有的方法通常依赖于所有模式的完整性,而在现实世界的应用中,由于噪声、硬件故障或隐私限制等问题,往往会出现缺失模式的情况。不完整模式与完整模式之间存在显著的功能偏差,直接融合它们甚至可能扭曲了完好模式的良好学习表示。为此,我们提出了PRLF(渐进式表征学习框架),旨在解决不确定的缺失模式情况下的多模态情感分析问题。 该框架引入了一个自适应模式可靠性估计器(AMRE),它根据识别置信度和费雪信息动态量化每个模式的可靠性,并确定主导模式。此外,逐步交互模块(ProgInteract)通过迭代地将其他模式与主导模式对齐来增强跨模态一致性并抑制噪声。 在CMU-MOSI、CMU-MOSEI和SIMS数据集上的广泛实验表明,在不同缺失情况(包括跨模式和内模式缺失)下,PRLF优于最先进的方法,展示了其鲁棒性和泛化能力。
https://arxiv.org/abs/2603.09111
By capturing the prevailing sentiment and market mood, textual data has become increasingly vital for forecasting commodity prices, particularly in metal markets. However, the effectiveness of lightweight, finetuned large language models (LLMs) in extracting predictive signals for aluminum prices, and the specific market conditions under which these signals are most informative, remains under-explored. This study generates monthly sentiment scores from English and Chinese news headlines (Reuters, Dow Jones Newswires, and China News Service) and integrates them with traditional tabular data, including base metal indices, exchange rates, inflation rates, and energy prices. We evaluate the predictive performance and economic utility of these models through long-short simulations on the Shanghai Metal Exchange from 2007 to 2024. Our results demonstrate that during periods of high volatility, Long Short-Term Memory (LSTM) models incorporating sentiment data from a finetuned Qwen3 model (Sharpe ratio 1.04) significantly outperform baseline models using tabular data alone (Sharpe ratio 0.23). Subsequent analysis elucidates the nuanced roles of news sources, topics, and event types in aluminum price forecasting.
通过捕捉当前的情绪和市场氛围,文本数据在预测商品价格方面变得越来越重要,特别是在金属市场上。然而,轻量级微调的大规模语言模型(LLMs)提取铝价的预测信号的有效性以及这些信号在特定市场条件下的信息价值仍然没有得到充分探索。这项研究生成了来自英文和中文新闻标题(路透社、道琼斯通讯社和中国新闻社)的月度情绪评分,并将它们与传统的表格数据集(包括基础金属指数、汇率、通胀率和能源价格)相结合。我们通过2007年至2024年在上海金属交易所进行的长短交易模拟来评估这些模型的预测性能和经济效用。我们的结果表明,在高波动性时期,结合来自微调后的Qwen3模型的情绪数据的长短时记忆(LSTM)模型表现显著优于仅使用表格数据作为基础的基准模型(夏普比率分别为1.04和0.23)。后续分析进一步阐明了新闻来源、主题和事件类型在铝价预测中的复杂作用。
https://arxiv.org/abs/2603.09085
Large language models (LLMs) increasingly operate as autonomous agents that reason over external APIs to perform complex tasks. However, their reliability and agreement remain poorly characterized. We present a unified benchmarking framework to quantify inter-LLM divergence, defined as the extent to which models differ in API discovery and ranking under identical tasks. Across 15 canonical API domains and 5 major model families, we measure pairwise and group-level agreement using set-, rank-, and consensus-based metrics including Average Overlap, Jaccard similarity, Rank-Biased Overlap, Kendall's tau, Kendall's W, and Cronbach's alpha. Results show moderate overall alignment (AO about 0.50, tau about 0.45) but strong domain dependence: structured tasks (Weather, Speech-to-Text) are stable, while open-ended tasks (Sentiment Analysis) exhibit substantially higher divergence. Volatility and consensus analyses reveal that coherence clusters around data-bound domains and degrades for abstract reasoning tasks. These insights enable reliability-aware orchestration in multi-agent systems, where consensus weighting can improve coordination among heterogeneous LLMs. Beyond performance benchmarking, our results reveal systematic failure modes in multi-agent LLM coordination, where apparent agreement can mask instability in action-relevant rankings. This hidden divergence poses a pre-deployment safety risk and motivates diagnostic benchmarks for early detection.
https://arxiv.org/abs/2604.22760
We present our system for SemEval-2026 Task 3 on dimensional aspect-based sentiment regression. Our approach combines a hybrid RoBERTa encoder, which jointly predicts sentiment using regression and discretized classification heads, with large language models (LLMs) via prediction-level ensemble learning. The hybrid encoder improves prediction stability by combining continuous and discretized sentiment representations. We further explore in-context learning with LLMs and ridge-regression stacking to combine encoder and LLM predictions. Experimental results on the development set show that ensemble learning significantly improves performance over individual models, achieving substantial reductions in RMSE and improvements in correlation scores. Our findings demonstrate the complementary strengths of encoder-based and LLM-based approaches for dimensional sentiment analysis. Our development code and resources will be shared at this https URL
我们介绍了我们的系统,用于参加2026年SemEval竞赛的第三项任务:基于维度的方面情感回归。我们的方法结合了一种混合RoBERTa编码器与大型语言模型(LLMs),通过预测级集成学习将两者结合起来。该混合编码器通过整合连续和离散的情感表示来提高预测稳定性,从而联合进行回归和离散分类以预测情感。我们进一步探索了在情境中使用LLM的在上下文学习以及岭回归堆叠技术,以便结合编码器和LLM的预测结果。 实验结果显示,在开发集上,集成学习显著提高了单一模型的表现力,实现了RMSE(均方根误差)的大幅度降低,并且相关得分有所提升。我们的研究发现表明了基于编码器的方法与基于大型语言模型的方法在维度情感分析中的互补优势。我们将通过以下链接分享我们的开发代码和资源:[此URL]
https://arxiv.org/abs/2603.07766
Stock market prediction presents considerable challenges for investors, financial institutions, and policymakers operating in complex market environments characterized by noise, non-stationarity, and behavioral dynamics. Traditional forecasting methods often fail to capture the intricate patterns and cross-sectional dependencies inherent in financial markets. This paper presents an integrated framework combining a node transformer architecture with BERT-based sentiment analysis for stock price forecasting. The proposed model represents the stock market as a graph structure where individual stocks form nodes and edges capture relationships including sectoral affiliations, correlated price movements, and supply chain connections. A fine-tuned BERT model extracts sentiment from social media posts and combines it with quantitative market features through attention-based fusion. The node transformer processes historical market data while capturing both temporal evolution and cross-sectional dependencies among stocks. Experiments on 20 S&P 500 stocks spanning January 1982 to March 2025 demonstrate that the integrated model achieves a mean absolute percentage error (MAPE) of 0.80% for one-day-ahead predictions, compared to 1.20% for ARIMA and 1.00% for LSTM. Sentiment analysis reduces prediction error by 10% overall and 25% during earnings announcements, while graph-based modeling contributes an additional 15% improvement by capturing inter-stock dependencies. Directional accuracy reaches 65% for one-day forecasts. Statistical validation through paired t-tests confirms these improvements (p < 0.05 for all comparisons). The model maintains MAPE below 1.5% during high-volatility periods where baseline models exceed 2%.
股票市场的预测对于投资者、金融机构和政策制定者来说在复杂市场环境中面临着巨大的挑战,这些环境充满了噪音、非平稳性和行为动态。传统预测方法往往无法捕捉到金融市场中固有的复杂模式和跨截面依赖性。本文提出了一种结合节点变压器架构与基于BERT的情感分析的综合框架,用于股票价格预测。 该模型将股票市场表示为图结构,在这种结构中,单个股票形成节点,边则代表包括行业关联、相关的价格变动以及供应链连接在内的关系。经过微调的BERT模型从社交媒体帖子中提取情感,并通过基于注意力机制的方法将其与定量市场特征相结合。节点变压器处理历史市场数据,同时捕捉股票之间的时序演变和跨截面依赖性。 在针对20只标普500成分股进行的实验中(涵盖1982年1月至2025年3月期间的数据),综合模型在一日前预测中的平均绝对百分比误差(MAPE)为0.80%,相比之下,ARIMA模型和LSTM模型分别为1.20%和1.00%。情感分析整体上减少了预测误差的10%,在财报发布期间减少幅度高达25%,而基于图的方法通过捕捉股票间的依赖关系贡献了额外15%的改进。方向准确性达到了65%的一日预报水平。配对t检验的统计验证证实了这些改进(所有比较中的p值均小于0.05)。在高波动性期间,模型维持MAPE低于1.5%,而基准模型则超过2%。 这一研究证明,在复杂且充满不确定性的金融市场中,综合使用先进的机器学习技术和基于图的方法可以显著提高预测准确性。
https://arxiv.org/abs/2603.05917
FreeTxt-Vi is a free and open source web based toolkit for creating and analysing bilingual Vietnamese English text collections. Positioned at the intersection of corpus linguistics and natural language processing NLP it enables users to build explore and interpret free text data without requiring programming expertise. The system combines corpus analysis features such as concordancing keyword analysis word relation exploration and interactive visualisation with transformer based NLP components for sentiment analysis and summarisation. A key contribution of this work is the design of a unified bilingual NLP pipeline that integrates a hybrid VnCoreNLP and Byte Pair Encoding BPE segmentation strategy a fine tuned TabularisAI sentiment classifier and a fine tuned Qwen2.5 model for abstractive summarisation. Unlike existing text analysis platforms FreeTxt Vi is evaluated as a set of language processing components. We conduct a three part evaluation covering segmentation sentiment analysis and summarisation and show that our approach achieves competitive or superior performance compared to widely used baselines in both Vietnamese and English. By reducing technical barriers to multilingual text analysis FreeTxt Vi supports reproducible research and promotes the development of language resources for Vietnamese a widely spoken but underrepresented language in NLP. The toolkit is applicable to domains including education digital humanities cultural heritage and the social sciences where qualitative text data are common but often difficult to process at scale.
FreeTxt-Vi 是一个免费且开源的基于网页的工具包,用于创建和分析越南语-英语双语文本集合。它位于语料库语言学与自然语言处理(NLP)的交叉点上,使用户能够在无需编程技能的情况下构建、探索并解释自由文本数据。该系统结合了如词汇共现分析、关键词分析、词关系探索以及交互式可视化等语料库分析功能,并集成了基于转换器的NLP组件以进行情感分析和摘要生成。 这项工作的关键贡献在于设计了一个统一的双语文本处理流水线,整合了一种混合策略(VnCoreNLP与字节对编码BPE分词),一个经过微调的TabularisAI情感分类器以及用于抽象式摘要生成的Qwen2.5模型。不同于现有的文本分析平台,FreeTxt-Vi作为一个语言处理组件集合进行评估。我们进行了三部分评估,涵盖分割、情感分析和总结,并展示了我们的方法在越南语与英语中均达到了相比广泛使用的基础线模型更为竞争或优越的表现。 通过降低多语言文本分析的技术门槛,FreeTxt-Vi支持可重复的研究并促进越南语资源的发展——这是一种被广泛使用的但又在NLP领域代表性不足的语言。该工具包适用于教育、数字人文、文化遗产和社科等领域,在这些领域中质性文本数据很常见但往往难以大规模处理。
https://arxiv.org/abs/2603.05690
In this paper, we present AILS-NTUA system for Track-A of SemEval-2026 Task 3 on Dimensional Aspect-Based Sentiment Analysis (DimABSA), which encompasses three complementary problems: Dimensional Aspect Sentiment Regression (DimASR), Dimensional Aspect Sentiment Triplet Extraction (DimASTE), and Dimensional Aspect Sentiment Quadruplet Prediction (DimASQP) within a multilingual and multi-domain framework. Our methodology combines fine-tuning of language-appropriate encoder backbones for continuous aspect-level sentiment prediction with language-specific instruction tuning of large language models using LoRA for structured triplet and quadruplet extraction. This unified yet task-adaptive design emphasizes parameter-efficient specialization across languages and domains, enabling reduced training and inference requirements while maintaining strong effectiveness. Empirical results demonstrate that the proposed models achieve competitive performance and consistently surpass the provided baselines across most evaluation settings.
在这篇论文中,我们介绍了AILS-NTUA系统,该系统用于SemEval-2026任务3的Track-A轨道,即维度基础情感分析(DimABSA)中的问题。这个问题包括三个互补的问题:维度属性情感回归(DimASR)、维度属性情感三元组抽取(DimASTE)以及维度属性情感四元组预测(DimASQP),并且这些问题在一个多语言和跨领域框架中进行研究。 我们的方法结合了针对连续属性级别的情感预测对适合各语言的编码器骨干网络进行微调,以及使用LoRA技术对大型语言模型进行特定语言指令调整以提取结构化的三元组和四元组。这种统一但任务适应性的设计强调在不同语言和领域中的参数效率专业化,从而减少训练和推理的需求,并保持强大的效果。 实证结果表明,所提出的模型达到了竞争性性能,在大多数评估设置中都超越了提供的基准方法。
https://arxiv.org/abs/2603.04933
Large Language Models (LLMs) often exhibit highly agreeable and reinforcing conversational styles, also known as AI-sycophancy. Although this behavior is encouraged, it may become problematic when interacting with user prompts that reflect negative social tendencies. Such responses risk amplifying harmful behavior rather than mitigating it. In this study, we examine how LLMs respond to user prompts expressing varying degrees of Dark Triad traits (Machiavellianism, Narcissism, and Psychopathy) using a curated dataset. Our analysis reveals differences across models, whereby all models predominantly exhibit corrective behavior, while showing reinforcing output in certain cases. Model behavior also depends on the severity level and differs in the sentiment of the response. Our findings raise implications for designing safer conversational systems that can detect and respond appropriately when users escalate from benign to harmful requests.
大型语言模型(LLMs)经常表现出高度顺从和强化的对话风格,也被称为人工智能奉迎。虽然这种行为在一定程度上是被鼓励的,但在与反映负面社会倾向的用户提示进行互动时可能会变得有问题。这样的回应风险在于放大有害行为而不是减轻它。在这项研究中,我们探讨了LLMs如何对表达不同程度黑暗三特质(马基雅维利主义、自恋和精神病态)的用户提示做出反应,使用了一个经过精心策划的数据集。我们的分析揭示了模型之间的差异,所有模型在大多数情况下都主要表现出纠正行为,但在某些情况下会显示出强化输出的行为。此外,模型的行为还取决于负面程度的不同,并且回应的情感色彩也不同。我们的研究结果提出了设计更安全的对话系统的重要性,这些系统能够检测并适当响应用户从良性到有害请求的升级情况。
https://arxiv.org/abs/2603.04299
Large Language Models (LLMs) are increasingly deployed in socially sensitive domains, yet their unpredictable behaviors, ranging from misaligned intent to inconsistent personality, pose significant risks. We introduce SteerEval, a hierarchical benchmark for evaluating LLM controllability across three domains: language features, sentiment, and personality. Each domain is structured into three specification levels: L1 (what to express), L2 (how to express), and L3 (how to instantiate), connecting high-level behavioral intent to concrete textual output. Using SteerEval, we systematically evaluate contemporary steering methods, revealing that control often degrades at finer-grained levels. Our benchmark offers a principled and interpretable framework for safe and controllable LLM behavior, serving as a foundation for future research.
大型语言模型(LLMs)在社会敏感领域中的部署日益增多,但它们不可预测的行为——从意图不一致到个性不统一——带来了显著的风险。我们引入了SteerEval,这是一个分层基准,用于评估LLM在其三个领域的可控性:语言特征、情感和个性。每个领域都分为三个规范层级:L1(要表达什么)、L2(如何表达)和L3(如何实例化),将高层次的行为意图与具体的文本输出联系起来。通过使用SteerEval,我们系统地评估了当前的导向方法,揭示出在更细粒度层次上控制往往有所下降。我们的基准测试提供了一个原则性和可解释性的框架,以实现安全且可控的LLM行为,并为未来的研究奠定了基础。
https://arxiv.org/abs/2603.02578
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.
我们提出了自洽结构化生成(Self-Consistent Structured Generation,SCSG)方法用于SemEval-2026任务3(Track A)中的维度方面情感分析。通过多次执行一个经过LoRA调整的大规模语言模型并仅保留多轮运行中达到多数共识的元组,SCSG增强了预测的可靠性。为了减少多次前向传递带来的计算开销,我们利用了vLLM的PagedAttention机制来高效地重复使用键值缓存。 跨6种语言和8种语言-领域组合进行的评估表明,自洽性通过15次执行可以显著提升统计学上优于单一推理提示的效果。我们的系统(利用Gemma 3)在所有设置中均排名前七,在四个英语子集中的三个获得第二名,并且在塔塔尔语餐厅数据集中于DimASTE任务中获得了第一名。
https://arxiv.org/abs/2603.01788
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.
训练面向方面的情感分析(ABSA)模型需要手动标注的数据,这既昂贵又耗时。本文介绍了一种名为LA-ABSA的新方法,该方法利用大型语言模型(LLM)生成的注释来微调轻量级模型,用于复杂的ABSA任务。我们在五个目标方面情感检测(TASD)和方面情感四元预测(ASQP)数据集上评估了我们的方法。我们的方法在增强策略中表现优异,并且在低资源场景下与LLM提示的表现相当,同时提供了显著的能源效率优势。例如,在使用50个标注示例进行上下文学习(ICL),以指导未标记数据的注释时,LA-ABSA在SemEval Rest16数据集上的ASQP任务上获得了49.85的F1分数,这一成绩几乎与使用Gemma-3-27B模型通过ICL提示得到的成绩(51.10)持平,但所需的计算资源却大幅减少。
https://arxiv.org/abs/2603.01778
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.
我们介绍了AnnoABSA,这是第一个支持全方位基于方面的情感分析(Aspect-Based Sentiment Analysis, ABSA)任务的网页注释工具。该工具高度可定制化,能够灵活配置情感元素和特定任务需求。除了手动标注外,AnnoABSA还提供了一种基于大型语言模型(Large Language Model, LLM)增强检索生成(Retrieval-Augmented Generation, RAG)建议的选项,在这种人机协作的方式中,可以为注释者提供上下文感知的帮助,并确保人类注释者的控制权。为了随着时间的推移提高预测质量,系统会检索已标注的最相似的十个例子并将它们作为提示中的少量示例添加进去,从而使得这些建议在注释过程中逐渐变得更加准确。AnnoABSA作为一个开源软件,在MIT许可证下发布,并且可以免费获取和轻松扩展以用于研究及实际应用中。
https://arxiv.org/abs/2603.01773
Multimodal sentiment analysis in federated learning environments faces significant challenges due to missing modalities, heterogeneous data distributions, and unreliable client updates. Existing federated approaches often struggle to maintain robust performance under these practical conditions. In this paper, we propose FedUAF, a unified multimodal federated learning framework that addresses these challenges through uncertainty-aware fusion and reliability-guided aggregation. FedUAF explicitly models modality-level uncertainty during local training and leverages client reliability to guide global aggregation, enabling effective learning under incomplete and noisy multimodal data. Extensive experiments on CMU-MOSI and CMU-MOSEI demonstrate that FedUAF consistently outperforms state-of-the-art federated baselines across various missing-modality patterns and Non-IID settings. Moreover, FedUAF exhibits superior robustness against noisy clients, highlighting its potential for real-world multimodal federated applications.
https://arxiv.org/abs/2603.13291
Large language models (LLMs) with reasoning capabilities have fueled a compelling narrative that reasoning universally improves performance across language tasks. We test this claim through a comprehensive evaluation of 504 configurations across seven model families--including adaptive, conditional, and reinforcement learning-based reasoning architectures--on sentiment analysis datasets of varying granularity (binary, five-class, and 27-class emotion). Our findings reveal that reasoning effectiveness is strongly task-dependent, challenging prevailing assumptions: (1) Reasoning shows task-complexity dependence--binary classification degrades up to -19.9 F1 percentage points (pp), while 27-class emotion recognition gains up to +16.0pp; (2) Distilled reasoning variants underperform base models by 3-18 pp on simpler tasks, though few-shot prompting enables partial recovery; (3) Few-shot learning improves over zero-shot in most cases regardless of model type, with gains varying by architecture and task complexity; (4) Pareto frontier analysis shows base models dominate efficiency-performance trade-offs, with reasoning justified only for complex emotion recognition despite 2.1x-54x computational overhead. We complement these quantitative findings with qualitative error analysis revealing that reasoning degrades simpler tasks through systematic over-deliberation, offering mechanistic insight beyond the high-level overthinking hypothesis.
具有推理能力的大规模语言模型(LLMs)推动了一种引人注目的观点,即推理在各种语言任务中普遍提高了性能。我们通过评估七类不同模型家族的504种配置——包括自适应、条件和基于强化学习的推理架构——在二元分类、五类以及27类情感识别等具有不同粒度的情感分析数据集上进行了全面测试,以验证这一观点。 我们的研究发现表明,推理的有效性强烈依赖于具体任务: 1. **任务复杂性的依赖关系**:对于简单的二元分类任务,性能下降最多可达到-19.9个F1百分点(pp),而对于复杂的27类情感识别任务,则可以提升高达+16.0pp。 2. **蒸馏推理变体的表现较基础模型在简单任务上表现较差**:这些变体通常比基础模型低3-18 pp,尽管通过少量样本提示可以在某种程度上恢复性能损失。 3. **少量示例学习的改进与零示例方法相比**:无论模型类型如何,在大多数情况下,少量示例学习都能超越零示例学习方法的表现,其提升量因架构和任务复杂性的不同而异。 4. **帕累托前沿分析显示基础模型在效率-性能权衡中占主导地位**:尽管推理带来了2.1倍至54倍的计算开销,但仅对于复杂的感情识别任务而言,引入推理是合理的。 此外,我们通过定性错误分析补充了这些定量发现,揭示出推理对简单任务的表现下降主要是由于过度思考(over-deliberation),这为关于高阶过度思考假设提供了一种超越表面的理解机制。
https://arxiv.org/abs/2602.24060
This paper introduces a novel Czech dataset in the restaurant domain for aspect-based sentiment analysis (ABSA), enriched with annotations of opinion terms. The dataset supports three distinct ABSA tasks involving opinion terms, accommodating varying levels of complexity. Leveraging this dataset, we conduct extensive experiments using modern Transformer-based models, including large language models (LLMs), in monolingual, cross-lingual, and multilingual settings. To address cross-lingual challenges, we propose a translation and label alignment methodology leveraging LLMs, which yields consistent improvements. Our results highlight the strengths and limitations of state-of-the-art models, especially when handling the linguistic intricacies of low-resource languages like Czech. A detailed error analysis reveals key challenges, including the detection of subtle opinion terms and nuanced sentiment expressions. The dataset establishes a new benchmark for Czech ABSA, and our proposed translation-alignment approach offers a scalable solution for adapting ABSA resources to other low-resource languages.
本文介绍了一个新颖的面向餐厅领域的基于方面的情感分析(ABSA)捷克语数据集,并辅以意见术语注释。该数据集支持三种涉及意见术语的不同难度级别的ABS任务。利用此数据集,我们使用现代Transformer模型(包括大规模语言模型(LLMs)),在单语、跨语和多语环境中进行了广泛的实验。为了解决跨语言挑战,我们提出了一种利用LLMs的翻译和标签对齐方法,并取得了持续改进的效果。我们的研究结果强调了当前最先进模型的优点和局限性,尤其是在处理像捷克语这样的低资源语言的复杂语言特性时。 详细错误分析揭示了一些关键挑战,包括检测细微的意见术语以及细腻的情感表达。该数据集为捷克语ABS领域建立了新的基准,并且我们提出的翻译对齐方法为将ABS资源适应其他低资源语言提供了一种可扩展的解决方案。
https://arxiv.org/abs/2602.22730
We introduce DLT-Corpus, the largest domain-specific text collection for Distributed Ledger Technology (DLT) research to date: 2.98 billion tokens from 22.12 million documents spanning scientific literature (37,440 publications), United States Patent and Trademark Office (USPTO) patents (49,023 filings), and social media (22 million posts). Existing Natural Language Processing (NLP) resources for DLT focus narrowly on cryptocurrencies price prediction and smart contracts, leaving domain-specific language under explored despite the sector's ~$3 trillion market capitalization and rapid technological evolution. We demonstrate DLT-Corpus' utility by analyzing technology emergence patterns and market-innovation correlations. Findings reveal that technologies originate in scientific literature before reaching patents and social media, following traditional technology transfer patterns. While social media sentiment remains overwhelmingly bullish even during crypto winters, scientific and patent activity grow independently of market fluctuations, tracking overall market expansion in a virtuous cycle where research precedes and enables economic growth that funds further innovation. We publicly release the full DLT-Corpus; LedgerBERT, a domain-adapted model achieving 23% improvement over BERT-base on a DLT-specific Named Entity Recognition (NER) task; and all associated tools and code.
我们介绍DLT-Corpus,这是迄今为止最大的专注于分布式账本技术(DLT)研究的特定领域文本集合:包含来自2212万份文档的298亿个词素,这些文档涵盖了科学文献(37,440篇出版物)、美国专利商标局(USPTO)专利申请(49,023项申请),以及社交媒体上的帖子(2200万个)。现有的自然语言处理(NLP)资源对DLT的研究重点狭隘地集中在加密货币价格预测和智能合约上,尽管该行业的市值约为3万亿美元,并且技术迅速演变,但特定领域的语言仍未得到充分探索。我们通过分析技术出现模式和市场创新相关性来展示DLT-Corpus的实用性。研究结果表明,技术首先出现在科学文献中,然后进入专利申请和社会媒体,遵循传统的技术转让模式。尽管在加密货币寒冬期间社交媒体情绪依然普遍乐观,但科学研究和技术专利活动却独立于市场波动,跟踪整体市场的扩张,在良性循环中促进经济发展,并为更多的创新提供资金支持。我们公开发布了完整的DLT-Corpus;LedgerBERT,这是一个针对DLT特定命名实体识别(NER)任务的领域适应模型,在该任务上比BERT-base模型提高了23%的性能;以及所有相关的工具和代码。
https://arxiv.org/abs/2602.22045
Implicit Discourse Relation Recognition (IDRR) remains a challenging task due to the requirement for deep semantic understanding in the absence of explicit discourse markers. A further limitation is that existing methods only predict relations without providing any supporting explanations. Recent advances in large language models (LLMs) have shown strong reasoning capabilities in both deep language understanding and natural language explanation generation. In this work, we propose a simple yet effective approach to distill the reasoning capabilities of LLMs into lightweight IDRR models to improve both performance and interpretability. Specifically, we first prompt an LLM to generate explanations for each training instance conditioned on its gold label. Then, we introduce a novel classification-generation framework that jointly performs relation prediction and explanation generation, and train it with the additional supervision of LLM-generated explanations. Our framework is plug-and-play, enabling easy integration with most existing IDRR models. Experimental results on PDTB demonstrate that our approach significantly improves IDRR performance, while human evaluation further confirms that the generated explanations enhance model interpretability. Furthermore, we validate the generality of our approach on sentiment classification and natural language inference
隐式话语关系识别(IDRR)仍然是一项具有挑战性的任务,因为它要求在缺乏明确的话语标记的情况下进行深度语义理解。现有方法的另一个限制是它们仅预测关系而不提供任何支持性解释。近年来,大型语言模型(LLMs)在深入的语言理解和自然语言解释生成方面表现出强大的推理能力。在这项工作中,我们提出了一种简单而有效的方法,即将LLM的推理能力提炼到轻量级IDRR模型中,以提高性能和可解释性。具体而言,我们首先提示一个LLM根据每个训练实例的黄金标签生成解释。然后,我们引入了一个新颖的分类-生成框架,该框架同时进行关系预测和解释生成,并通过附加的LLM生成的解释监督对其进行训练。我们的框架是即插即用型,能够轻松集成到大多数现有的IDRR模型中。 在PDTB数据集上的实验结果表明,我们的方法显著提高了IDRR性能,而人类评估进一步证实了所生成的解释增强了模型的可解释性。此外,我们还验证了该方法在情感分类和自然语言推理任务中的通用性。
https://arxiv.org/abs/2602.21763