While Large Language Models (LLMs) have demonstrated impressive performance across natural language generation tasks, their ability to generate truly creative content-characterized by novelty, diversity, surprise, and quality-remains limited. Existing methods for enhancing LLM creativity often focus narrowly on diversity or specific tasks, failing to address creativity's multifaceted nature in a generalizable way. In this work, we propose Creative Preference Optimization (CrPO), a novel alignment method that injects signals from multiple creativity dimensions into the preference optimization objective in a modular fashion. We train and evaluate creativity-augmented versions of several models using CrPO and MuCE, a new large-scale human preference dataset spanning over 200,000 human-generated responses and ratings from more than 30 psychological creativity assessments. Our models outperform strong baselines, including GPT-4o, on both automated and human evaluations, producing more novel, diverse, and surprising generations while maintaining high output quality. Additional evaluations on NoveltyBench further confirm the generalizability of our approach. Together, our results demonstrate that directly optimizing for creativity within preference frameworks is a promising direction for advancing the creative capabilities of LLMs without compromising output quality.

A trending line of recent work advocates for using large language models (LLMs) as formalizers instead of as end-to-end solvers for logical reasoning problems. Instead of generating the solution, the LLM generates a formal program that derives a solution via an external solver. While performance gain of the seemingly scalable LLM-as-formalizer over the seemingly unscalable LLM-as-solver has been widely reported, we show that this superiority does not hold on real-life constraint satisfaction problems. On 4 domains, we systematically evaluate 6 LLMs including 4 large reasoning models with inference-time scaling, paired with 5 pipelines including 2 types of formalism. We show that in few-shot settings, LLM-as-formalizer underperforms LLM-as-solver. While LLM-as-formalizer promises accuracy, robustness, faithfulness, and efficiency, we observe that the present LLMs do not yet deliver any of those, as their limited ability to generate formal programs leads to failure to scale with complexity, hard-coded solutions, and excessive reasoning tokens. We present our detailed analysis and actionable remedies to drive future research that improves LLM-as-formalizer.

We present StreamBridge, a simple yet effective framework that seamlessly transforms offline Video-LLMs into streaming-capable models. It addresses two fundamental challenges in adapting existing models into online scenarios: (1) limited capability for multi-turn real-time understanding, and (2) lack of proactive response mechanisms. Specifically, StreamBridge incorporates (1) a memory buffer combined with a round-decayed compression strategy, supporting long-context multi-turn interactions, and (2) a decoupled, lightweight activation model that can be effortlessly integrated into existing Video-LLMs, enabling continuous proactive responses. To further support StreamBridge, we construct Stream-IT, a large-scale dataset tailored for streaming video understanding, featuring interleaved video-text sequences and diverse instruction formats. Extensive experiments show that StreamBridge significantly improves the streaming understanding capabilities of offline Video-LLMs across various tasks, outperforming even proprietary models such as GPT-4o and Gemini 1.5 Pro. Simultaneously, it achieves competitive or superior performance on standard video understanding benchmarks.

Despite extensive efforts in safety alignment, large language models (LLMs) remain vulnerable to jailbreak attacks. Activation steering offers a training-free defense method but relies on fixed steering coefficients, resulting in suboptimal protection and increased false rejections of benign inputs. To address this, we propose AdaSteer, an adaptive activation steering method that dynamically adjusts model behavior based on input characteristics. We identify two key properties: Rejection Law (R-Law), which shows that stronger steering is needed for jailbreak inputs opposing the rejection direction, and Harmfulness Law (H-Law), which differentiates adversarial and benign inputs. AdaSteer steers input representations along both the Rejection Direction (RD) and Harmfulness Direction (HD), with adaptive coefficients learned via logistic regression, ensuring robust jailbreak defense while preserving benign input handling. Experiments on LLaMA-3.1, Gemma-2, and Qwen2.5 show that AdaSteer outperforms baseline methods across multiple jailbreak attacks with minimal impact on utility. Our results highlight the potential of interpretable model internals for real-time, flexible safety enforcement in LLMs.

Usability testing is a fundamental research method that user experience (UX) researchers use to evaluate and iterate their new designs. But what about evaluating and iterating the usability testing study design itself? Recent advances in Large Language Model-simulated Agent (LLM Agent) research inspired us to design UXAgent to support UX researchers in evaluating and iterating their study design before they conduct the real human-subject study. Our system features a Persona Generator module, an LLM Agent module, and a Universal Browser Connector module to automatically generate thousands of simulated users and to interactively test the target website. The system also provides a Result Viewer Interface so that the UX researchers can easily review and analyze the generated qualitative (e.g., agents' post-study surveys) and quantitative data (e.g., agents' interaction logs), or even interview agents directly. Through a heuristic evaluation with 16 UX researchers, participants praised the innovation of our system but also expressed concerns about the future of LLM Agent usage in UX studies.

Abstract--Fault diagnosis prevents train disruptions by ensuring the stability and reliability of their transmission systems. Data-driven fault diagnosis models have several advantages over traditional methods in terms of dealing with non-linearity, adaptability, scalability, and automation. However, existing data-driven models are trained on separate transmission components and only consider single faults due to the limitations of existing datasets. These models will perform worse in scenarios where components operate with each other at the same time, affecting each component's vibration signals. T o address some of these challenges, we propose a frequency domain representation and a 1-dimensional convolutional neural network for compound fault diagnosis and applied it on the PHM Beijing 2024 dataset, which includes 21 sensor channels, 17 single faults, and 42 compound faults from 4 interacting components, that is, motor, gearbox, left axle box, and right axle box. Our proposed model achieved 97.67% and 93.93% accuracies on the test set with 17 single faults and on the test set with 42 compound faults, respectively. Fault diagnosis plays a crucial role in maintaining the stability and reliability of transmission components, helping to prevent disruptions in train operations.

A fundamental technique lets researchers use a big, expensive model to train another model for less. The Chinese AI company DeepSeek released a chatbot earlier this year called R1, which drew a huge amount of attention. Most of it focused on the fact that a relatively small and unknown company said it had built a chatbot that rivaled the performance of those from the world's most famous AI companies, but using a fraction of the computer power and cost. As a result, the stocks of many Western tech companies plummeted; Nvidia, which sells the chips that run leading AI models, lost more stock value in a single day than any company in history. Some of that attention involved an element of accusation.

The Metaculus Cup required entrants to forecast the likelihood of 60 events over the summer. The Metaculus Cup required entrants to forecast the likelihood of 60 events over the summer. ManticAI ranked eighth in the Metaculus Cup, leaving some believing bots' prediction skills could soon overtake experts An artificial intelligence system has beaten scores of forecasting enthusiasts, including several professionals, in a contest to predict events ranging from bust-ups between Donald Trump and Elon Musk to Kemi Badenoch being removed from the Conservative party leadership. A British AI startup, co-founded by a former Google DeepMind researcher, has ranked in the top 10 of an international forecasting competition, which requires entrants to forecast the likelihood of 60 events over the summer. ManticAI came eighth in the Metaculus Cup, run by a San Francisco-based forecasting company that tries to predict the future for investment funds and corporations.

When you purchase through links in our articles, we may earn a small commission. Is Windows' Copilot button doomed to the fate of the Cortana key? From Cortana to Windows Copilot, history says new shortcut keys rarely stick. When Microsoft's Copilot key first poofed into existence, I tilted my head and thought . Does anyone remember the Cortana key?

On its 20th anniversary, YouTube is venturing into an era of AI-generated video, and may never be the same. Google figured out early on that video would be a great addition to its search business, so in 2005 it launched Google Video. Focused on making deals with the entertainment industry for second-rate content, and overly cautious on what users could upload, it flopped . In 2006, Google snapped up that year-old company, figuring it would sort out the IP stuff later. Though the $1.65 billion purchase price for YouTube was about a billion dollars more than its valuation, it was one of the greatest bargains ever.