As Large Language Models (LLMs) become more widely used for tasks like document summarization, question answering, and information extraction, improving their trustworthiness and interpretability has become increasingly important. One key strategy for achieving this is extbfattribution, a process that tracks the sources of the generated responses. This tutorial will explore various attribution techniques, including model-driven attribution, post-retrieval answering, and post-generation attribution. We will also discuss the challenges involved in implementing these approaches, and also look at the advanced topics such as model-based attribution for complex cases, table attribution, multimodal attribution, and multilingual attribution.

Continual learning (CL) enables deep learning models to learn a sequence of tasks under resource constraint settings, without forgetting previously acquired knowledge. This is particularly useful for multilingual NLP for low-resource languages, where incremental data collection is common and the compute cost is crucial. This tutorial will introduce key CL methodologies and their applications in natural language processing (NLP), covering both foundational techniques and modern challenges posed by large language models (LLMs). This tutorial covers foundational CL strategies based on regularization, replay, and network architecture. We explore NLP-specific CL scenarios such as task-incremental, language-incremental, and joint task-language incremental setups, along with methodologies to address them. A major emphasize of the tutorial is on continual learning for large language models (LLMs), examining challenges in applying CL for LLMs and the benefits it can provide in LLM training and inference. We further explore the connection between several advances in LLM such as model merging and continual learning. This tutorial is suitable for NLP researchers, practitioners, and students interested in lifelong learning, multilingual NLP, or large language models. It is designed as a half-day tutorial at IJCNLP 2025 and fall under the category of Introduction to Non-CL/Non-NLP Topic.

Affect refers to the fundamental neural processes that generate and regulate emotions, moods, and feeling states. Affect and emotions are central to how we organize meaning, to our behaviour, to our health and well-being, and to our very survival. Despite this, and even though most of us are all intimately familiar with emotions in everyday life, there is much we do not know about how emotions work, and how they impact our lives. Affective Science is a broad interdisciplinary field that explores these and other related questions about affect and emotions.Since language is a powerful mechanism of emotion expression, there is great potential in using language data and computation to shed light on fundamental questions about emotions. However, even though much progress has been made in areas such as sentiment analysis and affective computing, much of the research focus is squarely on automatically classifying pieces of text. In this tutorial, we will present an introduction to Affective Science and argue that NLP is uniquely positioned to contribute to it: to boldly explore a new frontier — to use language and computation to ask fundamental questions about how emotions and affect work. We will cover the broad areas of research within this nascent field of study - Computational Affective Science (CAS).

Human-agent teaming refers to humans and artificial agents working together toward shared goals, and recent advances in artificial intelligence, including large language models and autonomous robots, have intensified interest in using these agents not only for automation but also to augment higher-order cognition. Higher-order thinking involves complex mental processes such as critical thinking, creative problem solving, abstract reasoning, and metacognition, and intelligent agents hold the potential to act as genuine teammates that complement human strengths and address cognitive limitations. This tutorial synthesizes emerging research on human-agent teaming for cognitive augmentation by outlining the foundations of higher-order thinking and the psychological frameworks that describe it, reviewing key concepts and interaction paradigms in human–AI collaboration, and examining applications across education, healthcare, military decision-making, scientific discovery, and creative industries, where systems such as language models, decision-support tools, multi-agent architectures, explainable AI, and hybrid human–AI methods are used to support complex reasoning and expert judgment. It also discusses the major challenges involved in achieving meaningful augmentation, including the calibration of trust, the need for transparency, the development of shared mental models, the role of human adaptability and training, and broader ethical concerns. The tutorial further identifies gaps such as limited evidence of long-term improvement in human cognitive abilities and insufficient co-adaptation between humans and agents. Finally, it outlines future directions involving real-time cognitive alignment, long-term studies of cognitive development, co-adaptive learning systems, ethics-aware AI teammates, and new benchmarks for evaluating collaborative cognition, offering a comprehensive overview of current progress and a roadmap for advancing human-agent teaming as a means of enhancing higher-order human thinking.

LLMs are now embedded in workflows that span languages, modalities, and tools. This raises safety challenges that outpace conventional “guardrails”: jailbreaks and prompt injections, attributional safety failures under code-mixing, multimodal bypass via typography and icons, activation-level manipulation, and agentic risks from tool use. This tutorial synthesizes the newest advances (2023–2025) and lays out open research questions around (i) failure modes in monolingual / multilingual / multimodal settings, (ii) training-time and inference-time defenses (rejection SFT, RLHF/RLAIF, decoding-time safety, parameter/activation steering), and (iii) evaluation and red-teaming pipelines balancing safety and utility. We anchor the tutorial with recent results including our safety related papers published at top tier conferences, and connect them to emerging best practices from recent safety tutorials. The target audience is researchers/engineers with basic NLP knowledge who want the latest techniques and a research roadmap; format is half-day with short demos and Q&A.

This half-day tutorial provides a comprehensive overview of Legal Natural Language Processing (NLP) with LLM for participants with a basic understanding of Computational Linguistics or NLP concepts. We introduce how NLP can help analyze and manage legal text by covering five key topics: legal text analysis with LLM insights, legal text retrieval, rhetorical role identification, legal text summarization, and addressing bias and hallucination in legal tasks. Our goals are to explain why these tasks matter for researchers in the legal domain, describe the challenges and open problems, and outline current solutions. This proposed tutorial blends lectures, live examples, and Q&A to help researchers and students see how language technology and LLMs can make legal information more understandable and efficient.