We present ReasonIR-8B, the first retriever specifically trained for general reasoning tasks. Existing retrievers have shown limited gains on reasoning tasks, in part because existing training datasets focus on short factual queries tied to documents that straightforwardly answer them. We develop a synthetic data generation pipeline that, for each document, our pipeline creates a challenging and relevant query, along with a plausibly related but ultimately unhelpful hard negative. By training on a mixture of our synthetic data and existing public data, ReasonIR-8B achieves a new state-of-the-art of 29.9 nDCG@10 without reranker and 36.9 nDCG@10 with reranker on BRIGHT, a widely-used reasoning-intensive information retrieval (IR) benchmark. When applied to RAG tasks, ReasonIR-8B improves MMLU and GPQA performance by 6.4% and 22.6% respectively, relative to the closed-book baseline, outperforming other retrievers and search engines. In addition, ReasonIR-8B uses test-time compute more effectively: on BRIGHT, its performance consistently increases with longer and more information-rich rewritten queries; it continues to outperform other retrievers when combined with an LLM reranker. Our training recipe is general and can be easily extended to future LLMs; to this end, we open-source our code, data, and model.

This report addresses the challenge of limited labeled datasets for developing legal recommender systems, particularly in specialized domains like labor disputes. We propose a new approach leveraging the co-citation of legal articles within cases to establish similarity and enable algorithmic annotation. This method draws a parallel to the concept of case co-citation, utilizing cited articles as indicators of shared legal issues. To evaluate the labeled results, we employ a system that recommends similar cases based on plaintiffs' accusations, defendants' rebuttals, and points of disputes. The evaluation demonstrates that the recommender, with finetuned text embedding models and a reasonable BiLSTM module can recommend labor cases whose similarity was measured by the co-citation of the legal articles. This research contributes to the development of automated annotation techniques for legal documents, particularly in areas with limited access to comprehensive legal databases.

Ensuring that large language models (LLMs) are both helpful and harmless is a critical challenge, as overly strict constraints can lead to excessive refusals, while permissive models risk generating harmful content. Existing approaches, such as reinforcement learning from human feedback (RLHF) and direct preference optimization (DPO), attempt to balance these trade-offs but suffer from performance conflicts, limited controllability, and poor extendability. To address these issues, we propose Preference Vector, a novel framework inspired by task arithmetic. Instead of optimizing multiple preferences within a single objective, we train separate models on individual preferences, extract behavior shifts as preference vectors, and dynamically merge them at test time. This modular approach enables fine-grained, user-controllable preference adjustments and facilitates seamless integration of new preferences without retraining. Experiments show that our proposed Preference Vector framework improves helpfulness without excessive conservatism, allows smooth control over preference trade-offs, and supports scalable multi-preference alignment.

Personalized food recommendation systems (Food-RecSys) critically underperform due to fragmented component understanding and the failure of conventional machine learning with vast, imbalanced food data. While Large Language Models (LLMs) offer promise, current generic Recommendation as Language Processing (RLP) strategies lack the necessary specialization for the food domain's complexity. This thesis tackles these deficiencies by first identifying and analyzing the essential components for effective Food-RecSys. We introduce two key innovations: a multimedia food logging platform for rich contextual data acquisition and the World Food Atlas, enabling unique geolocation-based food analysis previously unavailable. Building on this foundation, we pioneer the Food Recommendation as Language Processing (F-RLP) framework - a novel, integrated approach specifically architected for the food domain. F-RLP leverages LLMs in a tailored manner, overcoming the limitations of generic models and providing a robust infrastructure for effective, contextual, and truly personalized food recommendations.

To responsibly develop Generative AI (GenAI) products, it is critical to define the scope of acceptable inputs and outputs. What constitutes a "safe" response is an actively debated question. Academic work puts an outsized focus on evaluating models by themselves for general purpose aspects such as toxicity, bias, and fairness, especially in conversational applications being used by a broad audience. In contrast, less focus is put on considering sociotechnical systems in specialized domains. Yet, those specialized systems can be subject to extensive and well-understood legal and regulatory scrutiny. These product-specific considerations need to be set in industry-specific laws, regulations, and corporate governance requirements. In this paper, we aim to highlight AI content safety considerations specific to the financial services domain and outline an associated AI content risk taxonomy. We compare this taxonomy to existing work in this space and discuss implications of risk category violations on various stakeholders. We evaluate how existing open-source technical guardrail solutions cover this taxonomy by assessing them on data collected via red-teaming activities. Our results demonstrate that these guardrails fail to detect most of the content risks we discuss.

The trend towards large language models (LLMs) for guardrailing against undesired behaviors is increasing and has shown promise for censoring user inputs. However, increased latency, memory consumption, hosting expenses and non-structured outputs can make their use prohibitive. In this work, we show that task-specific data generation can lead to fine-tuned classifiers that significantly outperform current state of the art (SoTA) while being orders of magnitude smaller. Secondly, we show that using a single model, \texttt{MultiTaskGuard}, that is pretrained on a large synthetically generated dataset with unique task instructions further improves generalization. Thirdly, our most performant models, \texttt{UniGuard}, are found using our proposed search-based model merging approach that finds an optimal set of parameters to combine single-policy models and multi-policy guardrail models. % On 7 public datasets and 4 guardrail benchmarks we created, our efficient guardrail classifiers improve over the best performing SoTA publicly available LLMs and 3$^{\text{rd}}$ party guardrail APIs in detecting unsafe and safe behaviors by an average F1 score improvement of \textbf{29.92} points over Aegis-LlamaGuard and \textbf{21.62} over \texttt{gpt-4o}, respectively. Lastly, our guardrail synthetic data generation process that uses custom task-specific guardrail poli

The Trump administration is working on changes to a Biden-era rule that would limit global access to AI chips, including possibly doing away with its splitting the world into tiers that help determine how many advanced semiconductors a country can obtain, three sources familiar with the matter said. The sources said the plans were still under discussion and warned they could change. But if enacted, removing the tiers could open the door to using U.S. chips as an even more powerful negotiating tool in trade talks. The regulation, which was issued in January, is aimed at dividing up access to the most advanced AI chips and controlling certain model weights in order to keep the most sophisticated computing power in the United States and among its allies, and away from China and other countries of concern.

Societal cognitive overload, driven by the deluge of inform ation and complexity in the AI age, poses a critical challenge to human well-being an d societal resilience. This paper argues that mitigating cognitive overload is not only essential for improving present-day life but also a crucial prerequisite fo r navigating the potential risks of advanced AI, including existential threats. W e exa mine how AI exacerbates cognitive overload through various mechanisms, incl uding information proliferation, algorithmic manipulation, automation anxiet ies, deregulation, and the erosion of meaning. The paper reframes the AI safety debate t o center on cognitive overload, highlighting its role as a bridge between near-te rm harms and long-term risks. It concludes by discussing potential institutional adaptations, research directions, and policy considerations that arise from adopti ng an overload-resilient perspective on human-AI alignment, suggesting pathways fo r future exploration rather than prescribing definitive solutions. W e stand at a precipice. Human societies are increasingly st ruggling to process the sheer volume and complexity of information in the digital age, a conditio n dramatically amplified by the rapid proliferation of artificial intelligence (AI). While Toffle r (1970) foresaw "future shock" from accelerating change and Eppler & Mengis (2004); Bawden & Robin son (2009) analyzed individual information overload, Byung-Chul Han, in his critique of ne oliberalism and technological domination (Han, 2017), argues that contemporary society faces a regime of technological domination that exploits and overwhelms the psyche. This exploitation and overwhelming of the psyche, now dramatically amplified by AI-driven information and comple xity, elevates information overload to a systemic crisis: societal cognitive overload .

Large language models and autonomous AI agents have evolved rapidly, resulting in a diverse array of evaluation benchmarks, frameworks, and collaboration protocols. However, the landscape remains fragmented and lacks a unified taxonomy or comprehensive survey. Therefore, we present a side-by-side comparison of benchmarks developed between 2019 and 2025 that evaluate these models and agents across multiple domains. In addition, we propose a taxonomy of approximately 60 benchmarks that cover general and academic knowledge reasoning, mathematical problem-solving, code generation and software engineering, factual grounding and retrieval, domain-specific evaluations, multimodal and embodied tasks, task orchestration, and interactive assessments. Furthermore, we review AI-agent frameworks introduced between 2023 and 2025 that integrate large language models with modular toolkits to enable autonomous decision-making and multi-step reasoning. Moreover, we present real-world applications of autonomous AI agents in materials science, biomedical research, academic ideation, software engineering, synthetic data generation, chemical reasoning, mathematical problem-solving, geographic information systems, multimedia, healthcare, and finance. We then survey key agent-to-agent collaboration protocols, namely the Agent Communication Protocol (ACP), the Model Context Protocol (MCP), and the Agent-to-Agent Protocol (A2A). Finally, we discuss recommendations for future research, focusing on advanced reasoning strategies, failure modes in multi-agent LLM systems, automated scientific discovery, dynamic tool integration via reinforcement learning, integrated search capabilities, and security vulnerabilities in agent protocols.

Artificial intelligence (AI) has emerged as a ubiquitous concept in numerous domains, including the legal system. AI has the potential to revolutionize the functioning of the judiciary and the dispensation of justice. Incorporating AI into the legal system offers the prospect of enhancing decision-making for judges, lawyers, and legal professionals, while concurrently providing the public with more streamlined, efficient, and cost-effective services. The integration of AI into the legal landscape offers manifold benefits, encompassing tasks such as document review, legal research, contract analysis, case prediction, and decision-making. By automating laborious and error-prone procedures, AI has the capacity to alleviate the burden associated with these arduous tasks. Consequently, courts around the world have begun embracing AI technology as a means to enhance the administration of justice. However, alongside its potential advantages, the use of AI in the judiciary poses a range of ethical challenges. These ethical quandaries must be duly addressed to ensure the responsible and equitable deployment of AI systems. This article delineates the principal ethical challenges entailed in employing AI within the judiciary and provides recommendations to effectively address these issues.