Aligning large language models (LLMs) to human preferences is challenging in domains where preference data is unavailable. We address the problem of learning reward models for such target domains by leveraging feedback collected from simpler source domains, where human preferences are easier to obtain. Our key insight is that, while domains may differ significantly, human preferences convey \emph{domain-agnostic} concepts that can be effectively captured by a reward model. We propose \method, a framework that trains domain-invariant reward models by optimizing a dual loss: a domain loss that minimizes the divergence between source and target distribution, and a source loss that optimizes preferences on the source domain. We show \method is a general approach that we evaluate and analyze across 4 distinct settings: (1) Cross-lingual transfer (accuracy: $0.621 \rightarrow 0.661$), (2) Clean-to-noisy (accuracy: $0.671 \rightarrow 0.703$), (3) Few-shot-to-full transfer (accuracy: $0.845 \rightarrow 0.920$), and (4) Simple-to-complex tasks transfer (correlation: $0.508 \rightarrow 0.556$). Our code, models and data are available at \url{https://github.com/portal-cornell/dial}.

With the widespread use of social media, user-generated content has surged on online platforms. When such content includes hateful, abusive, offensive, or cyberbullying behavior, it is classified as toxic speech, posing a significant threat to the online ecosystem's integrity and safety. While manual content moderation is still prevalent, the overwhelming volume of content and the psychological strain on human moderators underscore the need for automated toxic speech detection. Previously proposed detection methods often rely on large annotated datasets; however, acquiring such datasets is both costly and challenging in practice. To address this issue, we propose an uncertainty-guided firewall for toxic speech in few-shot scenarios, U-GIFT, that utilizes self-training to enhance detection performance even when labeled data is limited. Specifically, U-GIFT combines active learning with Bayesian Neural Networks (BNNs) to automatically identify high-quality samples from unlabeled data, prioritizing the selection of pseudo-labels with higher confidence for training based on uncertainty estimates derived from model predictions. Extensive experiments demonstrate that U-GIFT significantly outperforms competitive baselines in few-shot detection scenarios. In the 5-shot setting, it achieves a 14.92\% performance improvement over the basic model. Importantly, U-GIFT is user-friendly and adaptable to various pre-trained language models (PLMs). It also exhibits robust performance in scenarios with sample imbalance and cross-domain settings, while showcasing strong generalization across various language applications. We believe that U-GIFT provides an efficient solution for few-shot toxic speech detection, offering substantial support for automated content moderation in cyberspace, thereby acting as a firewall to promote advancements in cybersecurity.

In the fast-changing realm of information, the capacity to construct coherent timelines from extensive event-related content has become increasingly significant and challenging. The complexity arises in aggregating related documents to build a meaningful event graph around a central topic. This paper proposes CHRONOS - Causal Headline Retrieval for Open-domain News Timeline SummarizatiOn via Iterative Self-Questioning, which offers a fresh perspective on the integration of Large Language Models (LLMs) to tackle the task of Timeline Summarization (TLS). By iteratively reflecting on how events are linked and posing new questions regarding a specific news topic to gather information online or from an offline knowledge base, LLMs produce and refresh chronological summaries based on documents retrieved in each round. Furthermore, we curate Open-TLS, a novel dataset of timelines on recent news topics authored by professional journalists to evaluate open-domain TLS where information overload makes it impossible to find comprehensive relevant documents from the web. Our experiments indicate that CHRONOS is not only adept at open-domain timeline summarization, but it also rivals the performance of existing state-of-the-art systems designed for closed-domain applications, where a related news corpus is provided for summarization.

The extraordinary success of recent Large Language Models (LLMs) on a diverse array of tasks has led to an explosion of scientific and philosophical theorizing aimed at explaining how they do what they do. Unfortunately, disagreement over fundamental theoretical issues has led to stalemate, with entrenched camps of LLM optimists and pessimists often committed to very different views of how these systems work. Overcoming stalemate requires agreement on fundamental questions, and the goal of this paper is to address one such question, namely: is LLM behavior driven partly by representation-based information processing of the sort implicated in biological cognition, or is it driven entirely by processes of memorization and stochastic table look-up? This is a question about what kind of algorithm LLMs implement, and the answer carries serious implications for higher level questions about whether these systems have beliefs, intentions, concepts, knowledge, and understanding. I argue that LLM behavior is partially driven by representation-based information processing, and then I describe and defend a series of practical techniques for investigating these representations and developing explanations on their basis. The resulting account provides a groundwork for future theorizing about language models and their successors.

The evolution of agentic systems represents a significant milestone in artificial intelligence and modern software systems, driven by the demand for vertical intelligence tailored to diverse industries. These systems enhance business outcomes through adaptability, learning, and interaction with dynamic environments. At the forefront of this revolution are Large Language Model (LLM) agents, which serve as the cognitive backbone of these intelligent systems. In response to the need for consistency and scalability, this work attempts to define a level of standardization for Vertical AI agent design patterns by identifying core building blocks and proposing a \textbf{Cognitive Skills } Module, which incorporates domain-specific, purpose-built inference capabilities. Building on these foundational concepts, this paper offers a comprehensive introduction to agentic systems, detailing their core components, operational patterns, and implementation strategies. It further explores practical use cases and examples across various industries, highlighting the transformative potential of LLM agents in driving industry-specific applications.

Recent advancements in large language models (LLMs) based embedding models have established new state-of-the-art benchmarks for text embedding tasks, particularly in dense vector-based retrieval. However, these models predominantly focus on English, leaving multilingual embedding capabilities largely unexplored. To address this limitation, we present LUSIFER, a novel zero-shot approach that adapts LLM-based embedding models for multilingual tasks without requiring multilingual supervision. LUSIFER's architecture combines a multilingual encoder, serving as a language-universal learner, with an LLM-based embedding model optimized for embedding-specific tasks. These components are seamlessly integrated through a minimal set of trainable parameters that act as a connector, effectively transferring the multilingual encoder's language understanding capabilities to the specialized embedding model. Additionally, to comprehensively evaluate multilingual embedding performance, we introduce a new benchmark encompassing 5 primary embedding tasks, 123 diverse datasets, and coverage across 14 languages. Extensive experimental results demonstrate that LUSIFER significantly enhances the multilingual performance across various embedding tasks, particularly for medium and low-resource languages, without requiring explicit multilingual training data.

Traditional approaches relied heavily on lexicon-based methods, which involved predefined lists of biased terms and phrases. While This study investigates the several nuanced rationales for countering these methods provided a foundation, they often struggled with the the rise of political bias. We evaluate the performance of nuanced and context-dependent nature of political bias. More contemporary the Llama-3 (70B) language model on the Media Bias Identification techniques leveraging deep learning and large language Benchmark (MBIB), based on a novel prompting technique that models (LLMs) to capture subtler forms of bias and context-specific incorporates subtle reasons for identifying political leaning. Our variations in sentence-level text have been done [9]. Efforts have findings underscore the challenges of detecting political bias and been made to learn the factuality of reporting and bias, trying to categorize highlight the potential of transfer learning methods to enhance entire news media based upon different features collected future models. Through our framework, we achieve a comparable through its URLS, websites, Wikipedia page, Twitter account and performance with the supervised and fully fine-tuned ConvBERT many other factors [3].

Long-sequence causal reasoning seeks to uncover causal relationships within extended time series data but is hindered by complex dependencies and the challenges of validating causal links. To address the limitations of large-scale language models (e.g., GPT-4) in capturing intricate emotional causality within extended dialogues, we propose CauseMotion, a long-sequence emotional causal reasoning framework grounded in Retrieval-Augmented Generation (RAG) and multimodal fusion. Unlike conventional methods relying only on textual information, CauseMotion enriches semantic representations by incorporating audio-derived features-vocal emotion, emotional intensity, and speech rate-into textual modalities. By integrating RAG with a sliding window mechanism, it effectively retrieves and leverages contextually relevant dialogue segments, thus enabling the inference of complex emotional causal chains spanning multiple conversational turns. To evaluate its effectiveness, we constructed the first benchmark dataset dedicated to long-sequence emotional causal reasoning, featuring dialogues with over 70 turns. Experimental results demonstrate that the proposed RAG-based multimodal integrated approach, the efficacy of substantially enhances both the depth of emotional understanding and the causal inference capabilities of large-scale language models. A GLM-4 integrated with CauseMotion achieves an 8.7% improvement in causal accuracy over the original model and surpasses GPT-4o by 1.2%. Additionally, on the publicly available DiaASQ dataset, CauseMotion-GLM-4 achieves state-of-the-art results in accuracy, F1 score, and causal reasoning accuracy.

LG has rolled out its Gram laptop lineup in advance of CES 2025, confirming Intel's debut of the Arrow Lake platform in laptops and offering a mix of AI software that can run either locally or in the cloud. LG disclosed two versions of the LG Gram Pro (the 17Z90TR and 16Z90TS), as well as the LG gram Pro 2-in-1. LG is adding a new entry to its lineup as well: the Gram Book, an entry-level lightweight PC with a 1080p display and a cheaper price tag to boot. However, LG isn't saying what that price tag will be, or when these new devices will ship. LG's two new Gram Pros offer the choice of either integrated or discrete graphics.

That's why we've created the AI Hype Index--a simple, at-a-glance summary of everything you need to know about the state of the industry. More than 70 countries went to the polls in 2024. The good news is that this year of global elections turned out to be largely free from any major deepfake campaigns or AI manipulation. Instead we saw lots of AI slop: buff Trump, Elon as ultra-Chad, California as catastrophic wasteland. While some worry that development of large language models is slowing down, you wouldn't know it from the steady drumbeat of new products, features, and services rolling out from itty-bitty startups and massive incumbents alike.