Hao Li, Yizheng Sun, Viktor Schlegel, Kailai Yang, Riza Batista-Navarro, Goran Nenadic. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Kailai Yang, Xiao Liu, Lei Ji, Hao Li, Xiao Liang, Zhiwei Liu, Yeyun Gong, Peng Cheng, Mao Yang. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Misinformation is prevalent in various fields such as education, politics, health, etc., causing significant harm to society. However, current methods for cross-domain misinformation detection rely on effort- and resource-intensive fine-tuning and complex model structures. With the outstanding performance of LLMs, many studies have employed them for misinformation detection. Unfortunately, they focus on in-domain tasks and do not incorporate significant sentiment and emotion features (which we jointly call affect). In this paper, we propose RAEmoLLM, the first retrieval augmented (RAG) LLMs framework to address cross-domain misinformation detection using in-context learning based on affective information. RAEmoLLM includes three modules. (1) In the index construction module, we apply an emotional LLM to obtain affective embeddings from all domains to construct a retrieval database. (2) The retrieval module uses the database to recommend top K examples (text-label pairs) from source domain data for target domain contents. (3) These examples are adopted as few-shot demonstrations for the inference module to process the target domain content. The RAEmoLLM can effectively enhance the general performance of LLMs in cross-domain misinformation detection tasks through affect-based retrieval, without fine-tuning. We evaluate our framework on three misinformation benchmarks. Results show that RAEmoLLM achieves significant improvements compared to the other few-shot methods on three datasets, with the highest increases of 15.64%, 31.18%, and 15.73% respectively. This project is available at https://github.com/lzw108/RAEmoLLM.

We propose ELAINE (EngLish-jApanese-chINesE)-medLLM, a trilingual (English, Japanese, Chinese) large language model adapted for the bio-medical domain based on Llama-3-8B. The training dataset was carefully curated in terms of volume and diversity to adapt to the biomedical domain and endow trilingual capability while preserving the knowledge and abilities of the base model. The training follows 2-stage paths: continued pre-training and supervised fine-tuning (SFT). Our results demonstrate that ELAINE-medLLM exhibits superior trilingual capabilities compared to existing bilingual or multilingual medical LLMs without severely sacrificing the base model’s capability.

Recent advancements in LLM alignment leverage token-level supervisions to perform finegrained preference optimization.However, existing token-level alignment methods either optimize on all available tokens, which can be noisy and inefficient, or perform selective training with complex and expensive key token selection strategies.In this work, we propose Selective Preference Optimization (SePO), a novel selective alignment strategy that centers on efficient key token selection without requiring strong, fine-grained supervision signals.We prove the feasibility of Direct Preference Optimization (DPO) as token-level reward function estimators, which applies to any existing alignment datasets and enables costefficient token selection with small-scale model sizes and training data.We then train an oracle model with DPO on the target data and utilize the estimated reward function to score all tokens within the target dataset, where only the key tokens are selected to supervise the target policy model with a contrastive objective function.Extensive experiments on three public evaluation benchmarks show that SePO significantly outperforms competitive baseline methods by only optimizing on 30% key tokens with up to 60% reduction in GPU training hours.We also explore SePO as a new paradigm for weakto-strong generalization, showing that weak oracle models effectively supervise strong policy models with up to 16.8× more parameters.SePO also selects useful supervision signals from out-of-distribution data, alleviating the over-optimization problem.The project is open-sourced here.

A unified foundation model for medical time series—pretrained on open access and ethically reviewed medical corpora—offers the potential to reduce annotation burdens, minimize model customization, and enable robust transfer across clinical institutions, modalities, and tasks, particularly in data-scarce or privacy-constrained environments. However, existing time series foundation models struggle to handle medical time series data due to its inherent challenges, including irregular intervals, heterogeneous sampling rates, and frequent missingness. To address these challenges, we introduce MIRA, a unified foundation model specifically designed for medical time series forecasting. MIRA incorporates a Continuous-Time Rotary Positional Encoding that enables fine-grained modeling of variable time intervals, a frequency-specific mixture-of-experts layer that routes computation across latent frequency regimes to further promote temporal specialization, and a Continuous Dynamics Extrapolation Block based on Neural ODE that models the continuous trajectory of latent states, enabling accurate forecasting at arbitrary target timestamps. Pretrained on a large-scale and diverse medical corpus comprising over 454 billion time points collect from publicly available datasets, MIRA achieving reductions in forecasting errors by an average of 8% and 6% in out-of-distribution and in-distribution scenarios, respectively. We also introduce a comprehensive benchmark spanning multiple downstream clinical tasks, establishing a foundation for future research in medical time series modeling.

Mengxi Xiao, Qianqian Xie, Ziyan Kuang, Zhicheng Liu, Kailai Yang, Min Peng, Weiguang Han, Jimin Huang. Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2024.

The internet has brought both benefits and harms to society. A prime example of the latter is misinformation, including conspiracy theories, which flood the web. Recent advances in natural language processing, particularly the emergence of large language models (LLMs), have improved the prospects of accurate misinformation detection. However, most LLM-based approaches to conspiracy theory detection focus only on binary classification and fail to account for the important relationship between misinformation and affective features (i.e., sentiment and emotions). Driven by a comprehensive analysis of conspiracy text that reveals its distinctive affective features, we propose ConspEmoLLM, the first open-source LLM that integrates affective information and is able to perform diverse tasks relating to conspiracy theories. These tasks include not only conspiracy theory detection, but also classification of theory type and detection of related discussion (e.g., opinions towards theories). ConspEmoLLM is fine-tuned based on an emotion-oriented LLM using our novel ConDID dataset, which includes five tasks to support LLM instruction tuning and evaluation. We demonstrate that when applied to these tasks, ConspEmoLLM largely outperforms several open-source general domain LLMs and ChatGPT, as well as an LLM that has been fine-tuned using ConDID, but which does not use affective features. ConspEmoLLM can be easily applied to identify and classify conspiracy-related text in the real world. The work has been released at https://github.com/lzw108/ConspEmoLLM/.

Sentiment analysis and emotion detection are important research topics in natural language processing (NLP) and benefit many downstream tasks. With the widespread application of large language models (LLMs), researchers have started exploring the application of LLMs based on instruction-tuning in the field of sentiment analysis. However, these models only focus on single aspects of affective classification tasks (e.g. sentimental polarity or categorical emotions), and overlook the regression tasks (e.g. sentiment strength or emotion intensity), which leads to poor performance in downstream tasks. The main reason is the lack of comprehensive affective instruction tuning datasets and evaluation benchmarks, which cover various affective classification and regression tasks. Moreover, although emotional information is useful for downstream tasks, existing downstream datasets lack high-quality and comprehensive affective annotations. In this paper, we propose EmoLLMs, the first series of open-sourced instruction-following LLMs for comprehensive affective analysis based on fine-tuning various LLMs with instruction data, the first multi-task affective analysis instruction dataset (AAID) with 234K data samples based on 3 classification tasks and 2 regression tasks to support LLM instruction tuning, and a comprehensive affective evaluation benchmark (AEB) with 8 regression tasks and 6 classification tasks from various sources and domains to test the generalization ability of LLMs. We propose a series of EmoLLMs by fine-tuning LLMs with AAID to solve various affective instruction tasks. We compare our models with a variety of LLMs and sentiment analysis tools on AEB, where our models outperform all other open-sourced LLMs and sentiment analysis tools, and surpass ChatGPT and GPT-4 in most tasks, which shows that the series of EmoLLMs achieve the ChatGPT-level and GPT-4-level generalization capabilities on affective analysis tasks, and demonstrates our models can be used as affective annotation tools. This project is available at https://github.com/lzw108/EmoLLMs/.

LLMs have transformed NLP and shown promise in various fields, yet their potential in finance is underexplored due to a lack of comprehensive benchmarks, the rapid development of LLMs, and the complexity of financial tasks. In this paper, we introduce FinBen, the first extensive open-source evaluation benchmark, including 42 datasets spanning 24 financial tasks, covering eight critical aspects: information extraction (IE), textual analysis, question answering (QA), text generation, risk management, forecasting, decision-making, and bilingual (English and Spanish). FinBen offers several key innovations: a broader range of tasks and datasets, the first evaluation of stock trading, novel agent and Retrieval-Augmented Generation (RAG) evaluation, and two novel datasets for regulations and stock trading. Our evaluation of 21 representative LLMs, including GPT-4, ChatGPT, and the latest Gemini, reveals several key findings: While LLMs excel in IE and textual analysis, they struggle with advanced reasoning and complex tasks like text generation and forecasting. GPT-4 excels in IE and stock trading, while Gemini is better at text generation and forecasting. Instruction-tuned LLMs improve textual analysis but offer limited benefits for complex tasks such as QA. FinBen has been used to host the first financial LLMs shared task at the FinNLP-AgentScen workshop during IJCAI-2024, attracting 12 teams. Their novel solutions outperformed GPT-4, showcasing FinBen's potential to drive innovations in financial LLMs. All datasets and code are publicly available for the research community, with results shared and updated regularly on the Open Financial LLM Leaderboard.

Recent advancements in large language models (LLMs) focus on aligning to heterogeneous human expectations and values via multi-objective preference alignment. However, existing methods are dependent on the policy model parameters, which require high-cost repetition of their alignment algorithms for each new policy model, and they cannot expand to unseen objectives due to their static alignment objectives. In this work, we propose Meta-Objective Aligner (MetaAligner), the first policy-agnostic and generalizable method for multi-objective preference alignment. MetaAligner models multi-objective alignment into three stages: (1) dynamic objectives reformulation algorithm reorganizes traditional alignment datasets to supervise the model on performing flexible alignment across different objectives; (2) conditional weak-to-strong correction paradigm aligns the weak outputs of fixed policy models to approach strong outputs with higher preferences in the corresponding alignment objectives, enabling plug-and-play inferences on any policy models, which significantly reduces training costs and facilitates alignment on close-source policy models; (3) generalizable inference method flexibly adjusts target objectives by updating their text descriptions in the prompts, facilitating generalizable alignment to unseen objectives. Experimental results show that MetaAligner achieves significant and balanced improvements in multi-objective alignments on 10 state-of-the-art policy models, and saves up to 93.63% of GPU training hours compared to previous alignment methods. The model also effectively aligns unseen objectives, marking the first step towards generalizable multi-objective preference alignment.

Early suicidal ideation detection using social media is crucial for mental health surveillance. Simultaneously, emojis from the posts can help us better understand users' emotions and predict mental health conditions. However, research in emoji-based suicide analysis remains underexplored, with few resources available, which can restrict the development of studying emoji usage patterns among users with suicidal ideation. In this work, we build a derived suicide-related emoji dataset named SuicidEmoji, which contains 25k emoji posts (2,329 suicide-related posts and 22,722 posts for the control group users) filtered from about 1.3 million crawled Reddit data. To the best of our knowledge, SuicidEmoji is the first suicide-related emoji dataset. Based on SuicidEmoji, we propose two novel tasks: emoji-aware suicidal ideation detection and emoji prediction, for which we build two benchmark subdatasets from SuicidEmoji to evaluate the performance of advanced methods including pre-trained language models (PLMs) and large language models (LLMs). We analyze the experimental results of two PLMs and the highly capable LLMs, which reveal the significance and challenges of emoji-based suicide-related NLP tasks. The dataset is avaliable at https://github.com/TianlinZhang668/SuicidEmoji.

As an integral part of people's daily lives, social media is becoming a rich source for automatic mental health analysis. As traditional discriminative methods bear poor generalization ability and low interpretability, the recent large language models (LLMs) have been explored for interpretable mental health analysis on social media, which aims to provide detailed explanations along with predictions in zero-shot or few-shot settings. The results show that LLMs still achieve unsatisfactory classification performance in a zero-shot/few-shot manner, which further significantly affects the quality of the generated explanations. Domain-specific finetuning is an effective solution, but faces two critical challenges: 1) lack of high-quality training data. 2) no open-source foundation LLMs. To alleviate these problems, we formally model interpretable mental health analysis as a text generation task, and build the first multi-task and multi-source interpretable mental health instruction (IMHI) dataset with 105K data samples to support LLM instruction tuning and evaluation. The raw social media data are collected from 10 existing sources covering 8 mental health analysis tasks. We prompt ChatGPT with expert-designed few-shot prompts to obtain explanations. To ensure the reliability of the explanations, we perform strict automatic and human evaluations on the correctness, consistency, and quality of generated data. Based on the IMHI dataset and LLaMA2 foundation models, we train MentaLLaMA, the first open-source instruction-following LLM series for interpretable mental health analysis on social media. We evaluate MentaLLaMA and other advanced methods on the IMHI benchmark, the first holistic evaluation benchmark for interpretable mental health analysis. The results show that MentaLLaMA approaches state-of-the-art discriminative methods in correctness and generates human-level explanations. MentaLLaMA models also show strong generalizability to unseen tasks. The project is available at https://github.com/SteveKGYang/MentaLLaMA.

In Emotion Recognition in Conversations (ERC), the emotions of target utterances are closely dependent on their context. Therefore, existing works train the model to generate the response of the target utterance, which aims to recognise emotions leveraging contextual information. However, adjacent response generation ignores long-range dependencies and provides limited affective information in many cases. In addition, most ERC models learn a unified distributed representation for each utterance, which lacks interpretability and robustness. To address these issues, we propose aVAD-disentangledVariationalAutoEncoder (VAD-VAE), which first introduces a target utterance reconstruction task based on Variational Autoencoder, then disentangles three affect representations Valence-Arousal-Dominance (VAD) from the latent space. We also enhance the disentangled representations by introducing VAD supervision signals from a sentiment lexicon and minimising the mutual information between VAD distributions. Experiments show that VAD-VAE outperforms the state-of-the-art model on two datasets. Further analysis proves the effectiveness of each proposed module and the quality of disentangled VAD representations. The code is available athttps://github.com/SteveKGYang/VAD-VAE.

The context-aware emotional reasoning ability of AI systems, especially in conversations, is of vital importance in applications such as online opinion mining from social media and empathetic dialogue systems. Due to the implicit nature of conveying emotions in many scenarios, commonsense knowledge is widely utilized to enrich utterance semantics and enhance conversation modeling. However, most previous knowledge infusion methods perform empirical knowledge filtering and design highly customized architectures for knowledge interaction with the utterances, which can discard useful knowledge aspects and limit their generalizability to different knowledge sources. Based on these observations, we propose a Bipartite Heterogeneous Graph (BHG) method for enhancing emotional reasoning with commonsense knowledge. In BHG, the extracted context-aware utterance representations and knowledge representations are modeled as heterogeneous nodes. Two more knowledge aggregation node types are proposed to perform automatic knowledge filtering and interaction. BHG-based knowledge infusion can be directly generalized to multi-type and multi-grained knowledge sources. In addition, we propose a Multi-dimensional Heterogeneous Graph Transformer (MHGT) to perform graph reasoning, which can retain unchanged feature spaces and unequal dimensions for heterogeneous node types during inference to prevent unnecessary loss of information. Experiments show that BHG-based methods significantly outperform state-of-the-art knowledge infusion methods and show generalized knowledge infusion ability with higher efficiency. Further analysis proves that previous empirical knowledge filtering methods do not guarantee to provide the most useful knowledge information. Our code is available at: https://github.com/SteveKGYang/BHG.

The latest large language models (LLMs) such as ChatGPT, exhibit strong capabilities in automated mental health analysis.However, existing relevant studies bear several limitations, including inadequate evaluations, lack of prompting strategies, and ignorance of exploring LLMs for explainability.To bridge these gaps, we comprehensively evaluate the mental health analysis and emotional reasoning ability of LLMs on 11 datasets across 5 tasks.We explore the effects of different prompting strategies with unsupervised and distantly supervised emotional information.Based on these prompts, we explore LLMs for interpretable mental health analysis by instructing them to generate explanations for each of their decisions.We convey strict human evaluations to assess the quality of the generated explanations, leading to a novel dataset with 163 humanassessed explanations 1 .We benchmark existing automatic evaluation metrics on this dataset to guide future related works.According to the results, ChatGPT shows strong in-context learning ability but still has a significant gap with advanced task-specific methods.Careful prompt engineering with emotional cues and expertwritten few-shot examples can also effectively improve performance on mental health analysis.In addition, ChatGPT generates explanations that approach human performance, showing its great potential in explainable mental health analysis.

Depressive symptoms identification on social media aims to identify posts from social media expressing symptoms of depression. This can be beneficial for developing mental health support systems and for understanding the symptoms of depression. The Patient Health Questionnaire-9 (PHQ-9) is an instrument that healthcare professionals widely use to assess and monitor symptoms of depression. However, most existing models only consider capturing semantic information from posts, without considering PHQ-9 descriptive information related to symptoms. In addition, they are not devised to capture features that are specific to each symptom, especially in the case of multi-label symptoms identification. To tackle these challenges, we present a Span-based PHQ-aware and similarity contrastive network (SpanPHQ). We first adopt a novel span-based framework casting depressive symptoms identification task as a span-prediction problem. Then, we introduce context-aware and PHQ-aware self-guided cross-attention modules to enhance the model’s ability to consider both semantic contextual information and PHQ-9 descriptive information. Besides, a similarity contrastive learning is designed to effectively utilise the label information in identifying class-specific features. Our model is evaluated on two depressive symptoms identification datasets, i.e., the D2S dataset with 1,850 Twitter posts and the PRIMATE dataset with 2,000 Reddit posts. Moreover, our model achieves competitive performance compared to existing models on both datasets, with macro-F1 of 63.22%, 68.84%, micro-F1 of 73.34%, 75.92%, weighted-F1 of 72.86%, 76.65%, JacS of 69.94%, 63.82% and HamL of 0.0665, 0.1832 on these two datasets, respectively. The ablation study further provides evidence of the effectiveness of each module we proposed. Furthermore, we include visualisations and case studies verifying the ability of our model to learn PHQ information and its superior performance over existing baselines. Our work is expected to help the future identification and analysis of depressive symptoms on social media.

A key challenge for Emotion Recognition in Conversations (ERC) is to distinguish semantically similar emotions. Some works utilise Supervised Contrastive Learning (SCL) which uses categorical emotion labels as supervision signals and contrasts in high-dimensional semantic space. However, categorical labels fail to provide quantitative information between emotions. ERC is also not equally dependent on all embedded features in the semantic space, which makes the high-dimensional SCL inefficient. To address these issues, we propose a novel low-dimensional Supervised Cluster-level Contrastive Learning (SCCL) method, which first reduces the high-dimensional SCL space to a three-dimensional affect representation space Valence-Arousal-Dominance (VAD), then performs cluster-level contrastive learning to incorporate measurable emotion prototypes. To help modelling the dialogue and enriching the context, we leverage the pre-trained knowledge adapters to infuse linguistic and factual knowledge. Experiments show that our method achieves new state-of-the-art results with$69.81\%$on IEMOCAP,$65.7\%$on MELD, and$62.51\%$on DailyDialog datasets. The analysis also proves that the VAD space is not only suitable for ERC but also interpretable, with VAD prototypes enhancing its performance and stabilising the training of SCCL. In addition, the pre-trained knowledge adapters benefit the performance of the utterance encoder and SCCL. Our code is available at:https://github.com/SteveKGYang/SCCL

Stress and depression detection on social media aim at the analysis of stress and identification of depression tendency from social media posts, which provide assistance for the early detection of mental health conditions. Existing methods mainly model the mental states of the post speaker implicitly. They also lack the ability to mentalise for complex mental state reasoning. Besides, they are not designed to explicitly capture class-specific features. To resolve the above issues, we propose a mental state Knowledge–aware and Contrastive Network (KC-Net). In detail, we first extract mental state knowledge from a commonsense knowledge base COMET, and infuse the knowledge using Gated Recurrent Units (GRUs) to explicitly model the mental states of the speaker. Then we propose a knowledge–aware mentalisation module based on dot-product attention to accordingly attend to the most relevant knowledge aspects. A supervised contrastive learning module is also utilised to fully leverage label information for capturing class-specific features. We test the proposed methods on a depression detection dataset Depression_Mixed with 3165 Reddit and blog posts, a stress detection dataset Dreaddit with 3553 Reddit posts, and a stress factors recognition dataset SAD with 6850 SMS-like messages. The experimental results show that our method achieves new state-of-the-art results on all datasets: 95.4% of F1 scores on Depression_Mixed, 83.5% on Dreaddit and 77.8% on SAD, with 2.07% average improvement. Factor-specific analysis and ablation study prove the effectiveness of all proposed modules, while UMAP analysis and case study visualise their mechanisms. We believe our work facilitates detection and analysis of depression and stress on social media data, and shows potential for applications on other mental health conditions.