We introduce Nemotron Nano V2 VL, the latest model of the Nemotron vision-language series designed for strong real-world document understanding, long video comprehension, and reasoning tasks. Nemotron Nano V2 VL delivers significant improvements over our previous model, Llama-3.1-Nemotron-Nano-VL-8B, across all vision and text domains through major enhancements in model architecture, datasets, and training recipes. Nemotron Nano V2 VL builds on Nemotron Nano V2, a hybrid Mamba-Transformer LLM, and innovative token reduction techniques to achieve higher inference throughput in long document and video scenarios. We are releasing model checkpoints in BF16, FP8, and FP4 formats and sharing large parts of our datasets, recipes and training code.

AI researchers have long focused on poker-like games as a testbed for environments characterized by multi-player dynamics, imperfect information, and reasoning under uncertainty. While recent breakthroughs have matched elite human play at no-limit Texas hold'em, the multi-player dynamics are subdued: most hands converge quickly with only two players engaged through multiple rounds of bidding. In this paper, we present Solly, the first AI agent to achieve elite human play in reduced-format Liar's Poker, a game characterized by extensive multi-player engagement. We trained Solly using self-play with a model-free, actor-critic, deep reinforcement learning algorithm. Solly played at an elite human level as measured by win rate (won over 50% of hands) and equity (money won) in heads-up and multi-player Liar's Poker. Solly also outperformed large language models (LLMs), including those with reasoning abilities, on the same metrics. Solly developed novel bidding strategies, randomized play effectively, and was not easily exploitable by world-class human players.

Configuration tuning is critical for database performance. Although recent advancements in database tuning have shown promising results in throughput and latency improvement, challenges remain. First, the vast knob space makes direct optimization unstable and slow to converge. Second, reinforcement learning pipelines often lack effective warm-start guidance and require long offline training. Third, transferability is limited: when hardware or workloads change, existing models typically require substantial retraining to recover performance. To address these limitations, we propose L2T-Tune, a new LLM-guided hybrid database tuning framework that features a three-stage pipeline: Stage one performs a warm start that simultaneously generates uniform samples across the knob space and logs them into a shared pool; Stage two leverages a large language model to mine and prioritize tuning hints from manuals and community documents for rapid convergence. Stage three uses the warm-start sample pool to reduce the dimensionality of knobs and state features, then fine-tunes the configuration with the Twin Delayed Deep Deterministic Policy Gradient algorithm. We conduct experiments on L2T-Tune and the state-of-the-art models. Compared with the best-performing alternative, our approach improves performance by an average of 37.1% across all workloads, and by up to 73% on TPC-C. Compared with models trained with reinforcement learning, it achieves rapid convergence in the offline tuning stage on a single server. Moreover, during the online tuning stage, it only takes 30 steps to achieve best results.

Generative AI (GenAI) has recently emerged as a groundbreaking force in Software Engineering, capable of generating code, identifying bugs, recommending fixes, and supporting quality assurance. While its use in coding tasks shows considerable promise, applying GenAI across the entire Software Development Life Cycle (SDLC) has not yet been fully explored. Critical uncertainties in areas such as reliability, accountability, security, and data privacy demand deeper investigation and coordinated action. The GENIUS project, comprising over 30 European industrial and academic partners, aims to address these challenges by advancing AI integration across all SDLC phases. It focuses on GenAI's potential, the development of innovative tools, and emerging research challenges, actively shaping the future of software engineering. This vision paper presents a shared perspective on the future of GenAI-driven software engineering, grounded in cross-sector dialogue as well as experiences and findings within the GENIUS consortium. The paper explores four central elements: (1) a structured overview of current challenges in GenAI adoption across the SDLC; (2) a forward-looking vision outlining key technological and methodological advances expected over the next five years; (3) anticipated shifts in the roles and required skill sets of software professionals; and (4) the contribution of GENIUS in realising this transformation through practical tools and industrial validation. This paper focuses on aligning technical innovation with business relevance. It aims to inform both research agendas and industrial strategies, providing a foundation for reliable, scalable, and industry-ready GenAI solutions for software engineering teams.

Low-Rank Adaptation (LoRA) has become a popular technique for parameter-efficient fine-tuning of large language models (LLMs). In many real-world scenarios, multiple adapters are loaded simultaneously to enable LLM customization for personalized user experiences or to support a diverse range of tasks. Although each adapter is lightweight in isolation, their aggregate cost becomes substantial at scale. This makes it possible to quantize the important components to higher precision, while quantizing the rest to ultra-low bitwidth. We conduct comprehensive experiments with LLaMA 2-7B, LLaMA 2-13B, and Mistral 7B models on mathematical reasoning, coding, and summarization tasks. Large Language Models (LLMs) have achieved remarkable performance across a wide range of natural language tasks (Ouyang et al., 2022; Wang et al., 2022; Zhao et al., 2023), but fine-tuning LLMs for new applications remains computationally and memory intensive. To address this challenge, low-rank adaptation (LoRA; Hu et al., 2022) has emerged as a widely adopted method for parameter-efficient fine-tuning. LoRA introduces small, task-specific low-rank matrices, and during the adaptation, only these low-rank matrices are trained while the base model is frozen. An increasingly important use case of LoRA is LLM customization, as LLM providers (e.g., OpenAI and Google) allow users to personalize their own LLMs (OpenAI, 2025; Google Cloud, 2025).

Large Language Models (LLMs) excel as passive responders, but teaching them to be proactive, goal-oriented partners--a critical capability in high-stakes domains--remains a major challenge. Current paradigms either myopically optimize single-turn attributes or rely on brittle, high-cost user simulators, creating a persistent "reality gap". To bridge this gap, we introduce Learn-to-Ask, a general, simulator-free framework for learning and deploying proactive dialogue agents directly from offline expert data, bypassing the need to model complex user dynamics. Our key insight is to reframe the offline policy learning problem by leveraging the observed future of each expert trajectory. This allows us to infer a dense, turn-by-turn reward signal grounded in the expert's revealed strategy, decomposing the intractable long-horizon problem into a series of supervised learning tasks, and training a policy to output a structured (action, state assessment) tuple, governing both what to ask and, crucially, when to stop. To ensure reward fidelity, our Automated Grader Calibration pipeline systematically purges noise from the LLM-based reward model with minimal human supervision. Empirically, we demonstrate the efficacy of Learn-to-Ask in a real-world medical dataset, using LLMs of varying sizes up to 32B. Our approach culminates in the successful deployment of LLMs into a live, large-scale online AI service. In rigorous in-house evaluations, our model was launched and achieved performance even superior to human experts, proving our framework's ability to translate offline data into tangible, real-world impact. We hope this work provides a practical and economically viable blueprint for transforming passive LLMs into proactive, goal-oriented LLM applications. Across industries such as healthcare, law, and finance, numerous goal-oriented conversations take place every day between human experts and their clients (Wang et al., 2025; Y ang et al., 2023). This vast corpus of dialogue data represents a largely untapped goldmine, containing implicit expert-driven strategies for navigating complex, information-seeking scenarios. While organizations possess these valuable data assets, Large Language Models (LLMs) are seldom trained to harness them effectively. Instead, their default behavior remains largely passive, limiting their potential as truly collaborative and proactive partners. In high-stakes domains, this passivity is a critical failure - an intelligent LLM application should not merely answer questions but proactively form a policy to gather information and drive the conversation towards a designated goal. Two main paradigms have emerged to instill such proactivity, yet both struggle with a significant "reality gap". It optimizes for local attributes and fails to learn a coherent, sequential policy that accounts for temporal dependencies in a conversation.

AI agents are continually optimized for tasks related to human work, such as software engineering and professional writing, signaling a pressing trend with significant impacts on the human workforce. However, these agent developments have often not been grounded in a clear understanding of how humans execute work, to reveal what expertise agents possess and the roles they can play in diverse workflows. In this work, we study how agents do human work by presenting the first direct comparison of human and agent workers across multiple essential work-related skills: data analysis, engineering, computation, writing, and design. To better understand and compare heterogeneous computer-use activities of workers, we introduce a scalable toolkit to induce interpretable, structured workflows from either human or agent computer-use activities. Using such induced workflows, we compare how humans and agents perform the same tasks and find that: (1) While agents exhibit promise in their alignment to human workflows, they take an overwhelmingly programmatic approach across all work domains, even for open-ended, visually dependent tasks like design, creating a contrast with the UI-centric methods typically used by humans. (2) Agents produce work of inferior quality, yet often mask their deficiencies via data fabrication and misuse of advanced tools. (3) Nonetheless, agents deliver results 88.3% faster and cost 90.4-96.2% less than humans, highlighting the potential for enabling efficient collaboration by delegating easily programmable tasks to agents.

We introduce Ling 2.0, a series reasoning-oriented language foundation built upon the principle that every activation boosts reasoning capability. Designed to scale from tens of billions to one trillion parameters under a unified Mixture-of-Experts (MoE) paradigm, Ling 2.0 emphasizes high sparsity, cross-scale consistency, and efficiency guided by empirical scaling laws. The series includes three non-thinking (instruct) models - Ling-mini-2.0, Ling-flash-2.0, and Ling-1T - ranging from 16B to 1T total parameters and achieving up to 7-fold active-compute efficiency compared with dense counterparts. Ling 2.0 integrates coordinated innovations across model architecture, pre-training, post-training, and infrastructure: a high-sparsity MoE with MTP for efficient reasoning, reasoning-oriented data and mid-training CoT activation, reinforcement-based fine-tuning (DFT, Evo-CoT), and full-scale FP8 training with fine-grained heterogeneous pipelines. At the trillion scale, Ling-1T establishes a new Pareto frontier of reasoning accuracy versus computational efficiency, demonstrating that sparse activation, when properly aligned with reasoning objectives, enables scalable and efficient intelligence. Collectively, Ling 2.0 provides a coherent, open, and efficient foundation for advancing future reasoning and thinking models, including the Ring series built upon the same base.

We introduce String Seed of Thought (SSoT), a novel prompting method for LLMs that improves Probabilistic Instruction Following (PIF). We define PIF as a task requiring an LLM to select its answer from a predefined set of options, each associated with a specific probability, such that the empirical distribution of the generated answers aligns with the target distribution when prompted multiple times. While LLMs excel at tasks with single, deterministic answers, they often fail at PIF, exhibiting biases problematic for applications requiring non-deterministic behaviors, such as human-behavior simulation, content diversification, and multiplayer games. It also harms the diversity of generated responses, a crucial factor in test-time scaling, by causing the outputs to collapse into a limited set of answers. To address this, we propose SSoT, a simple prompting method that instructs an LLM to first output a random string to generate sufficient entropy. SSoT also instructs the LLM to extract randomness by manipulating this string to derive a final answer, thereby preserving diversity while adhering to specific constraints. We demonstrate that SSoT significantly improves the PIF performance of LLMs, approaching the ideal performance of a pseudo-random number generator. Furthermore, our experiments on NoveltyBench show SSoT's benefits extend beyond closed-set tasks to open-ended tasks by enhancing response diversity.

Simulating human reasoning in open-ended tasks has long been a central aspiration in AI and cognitive science. While large language models now approximate human responses at scale, they remain tuned to population-level consensus, often erasing the individuality of reasoning styles and belief trajectories. To advance the vision of more human-like reasoning in machines, we introduce HugAgent (Human-Grounded Agent Benchmark), which rethinks human reasoning simulation along three dimensions: (i) from averaged to individualized reasoning, (ii) from behavioral mimicry to cognitive alignment, and (iii) from vignette-based to open-ended data. The benchmark evaluates whether a model can predict a specific person's behavioral responses and the underlying reasoning dynamics in out-of-distribution scenarios, given partial evidence of their prior views. HugAgent adopts a dual-track design: a human track that automates and scales the think-aloud method to collect ecologically valid human reasoning data, and a synthetic track for further scalability and systematic stress testing. This architecture enables low-cost, extensible expansion to new tasks and populations. Experiments with state-of-the-art language models reveal persistent adaptation gaps, positioning HugAgent as the first extensible benchmark for aligning machine reasoning with the individuality of human thought. The benchmark, along with its complete data collection pipeline and companion chatbot, is open-sourced as HugAgent (https://anonymous.4open.science/r/HugAgent) and TraceYourThinking (https://anonymous.4open.science/r/trace-your-thinking).