Dear reviewers, thank you for your time to thoroughly read and review our paper. Y ou said "the addressed problem is relevant and timely" (R4), "useful in practice with many applications" (R1), with Quantifying for the first time the gap between train and test losses in the approach of Ballé et al. (see below) The proposed approach only marginally improves PSNR for hyperprior models. Note that in compression lightweight models are very relevant in practice . Secondly, our empirical results allow us for the first time to quantify the gap between training and test losses . Finally, universal quantization has the potential to lead to much bigger improvements in the future .

Patients who are at clinical high risk (CHR) for schizophrenia need close monitoring of their symptoms to inform appropriate treatments. The Brief Psychiatric Rating Scale (BPRS) is a validated, commonly used research tool for measuring symptoms in patients with schizophrenia and other psychotic disorders; however, it is not commonly used in clinical practice as it requires a lengthy structured interview. Here, we utilize large language models (LLMs) to predict BPRS scores from clinical interview transcripts in 409 CHR patients from the Accelerating Medicines Partnership Schizophrenia (AMP-SCZ) cohort. Despite the interviews not being specifically structured to measure the BPRS, the zero-shot performance of the LLM predictions compared to the true assessment (median concordance: 0.84, ICC: 0.73) approaches human inter- and intra-rater reliability. We further demonstrate that LLMs have substantial potential to improve and standardize the assessment of CHR patients via their accuracy in assessing the BPRS in foreign languages (median concordance: 0.88, ICC: 0.70), and integrating longitudinal information in a one-shot or few-shot learning approach.

The large spread of disinformation across digital platforms creates significant challenges to information integrity. This paper presents a multi-agent system that uses relation extraction to detect disinformation in news articles, focusing on titles and short text snippets. The proposed Agentic AI system combines four agents: (i) a machine learning agent (logistic regression), (ii) a Wikipedia knowledge check agent (which relies on named entity recognition), (iii) a coherence detection agent (using LLM prompt engineering), and (iv) a web-scraped data analyzer that extracts relational triplets for fact checking. The system is orchestrated via the Model Context Protocol (MCP), offering shared context and live learning across components. Results demonstrate that the multi-agent ensemble achieves 95.3% accuracy with an F1 score of 0.964, significantly outperforming individual agents and traditional approaches. The weighted aggregation method, mathematically derived from individual agent misclassification rates, proves superior to algorithmic threshold optimization. The modular architecture makes the system easily scalable, while also maintaining details of the decision processes.

Can you imagine yourself having a conversation with Windows about what your PC is doing? Microsoft's Windows chief can, and is trying to build a future where those interactions are the norm. In an interview with Microsoft AI product manager Christiaan Brinkhoff, the chief of Microsoft's Windows Devices group, Pavan Davuluri, explained that the company is trying to work toward a future where you can access Windows pretty much anywhere via the cloud, then use AI to fine-tune what you're trying to accomplish. Microsoft described the conversation as "the next chapter of Windows," with an eye toward delivering the changes within the next few years. Davuluri described what he hoped the Windows team could accomplish from a strategic level, without targeting any future version of Windows with these goals in mind.

"Noise" protection can be added to content before it's uploaded online. A new technique developed by Australian researchers could stop unauthorised artificial intelligence (AI) systems learning from photos, artwork and other image-based content. Developed by CSIRO, Australia's national science agency, in partnership with the Cyber Security Cooperative Research Centre (CSCRC) and the University of Chicago, the method subtly alters content to make it unreadable to AI models while remaining unchanged to the human eye. This could help artists, organisations and social media users protect their work and personal data from being used to train AI systems or create deepfakes. For example, a social media user could automatically apply a protective layer to their photos before posting, preventing AI systems from learning facial features for deepfake creation.

Reasoning-augmented search agents such as Search-R1, trained via reinforcement learning with verifiable rewards (RLVR), demonstrate remarkable capabilities in multi-step information retrieval from external knowledge sources. These agents address the limitations of their parametric memory by dynamically gathering relevant facts to address complex reasoning tasks. However, existing approaches suffer from a fundamental architectural limitation: they process search queries strictly sequentially, even when handling inherently parallelizable and logically independent comparisons. This sequential bottleneck significantly constrains computational efficiency, particularly for queries that require multiple entity comparisons. To address this critical limitation, we propose ParallelSearch, a novel reinforcement learning framework that empowers large language models (LLMs) to recognize parallelizable query structures and execute multiple search operations concurrently. Our approach introduces dedicated reward functions that incentivize the identification of independent query components while preserving answer accuracy through jointly considering correctness, query decomposition quality, and parallel execution benefits. Comprehensive experiments demonstrate that ParallelSearch outperforms state-of-the-art baselines by an average performance gain of 2.9% across seven question-answering benchmarks. Notably, on parallelizable questions, our method achieves a 12.7% performance improvement while requiring only 69.6% of the LLM calls compared to sequential approaches.

We introduce MapStory, an LLM-powered animation prototyping tool that generates editable map animation sequences directly from natural language text by leveraging a dual-agent LLM architecture. Given a user written script, MapStory automatically produces a scene breakdown, which decomposes the text into key map animation primitives such as camera movements, visual highlights, and animated elements. Our system includes a researcher agent that accurately queries geospatial information by leveraging an LLM with web search, enabling automatic extraction of relevant regions, paths, and coordinates while allowing users to edit and query for changes or additional information to refine the results. Additionally, users can fine-tune parameters of these primitive blocks through an interactive timeline editor. We detail the system's design and architecture, informed by formative interviews with professional animators and by an analysis of 200 existing map animation videos. Our evaluation, which includes expert interviews (N=5) and a usability study (N=12), demonstrates that MapStory enables users to create map animations with ease, facilitates faster iteration, encourages creative exploration, and lowers barriers to creating map-centric stories.

AI-generated text is proliferating across domains, from creative writing and journalism to marketing content and scientific articles. Models can follow user-provided instructions to generate coherent and grammatically correct outputs but in this work, we study a more fundamental question: how do we evaluate and improve the writing quality of AI-generated text? Writing quality assessment has received less attention from the community, in part because it is fundamentally subjective and requires expertise. We first introduce the Writing Quality Benchmark (WQ) by consolidating five writing-preference datasets into 4,729 writing quality judgments. Our experiments show that most of the competitive baselines, including state-of-the-art LLMs that excel at reasoning tasks, barely outperform random baselines on WQ. We then train specialized Writing Quality Reward Models (WQRM) of various sizes for writing quality assessment that demonstrate strong generalization on four out-of-distribution test sets and 74% accuracy on the WQ benchmark. To further show WQRM's practical benefits during inference, we leverage additional test-time compute to generate and rank multiple candidate revisions, allowing us to select higher-quality outputs from an initial draft. Human evaluation with 9 experienced writers confirm that WQRM-based selection produces writing samples preferred by experts 66% overall, and 72.2% when the reward gap is larger than 1 point. We release our datasets and models to encourage community engagement with writing quality assessment and development of AI writing systems better aligned with human preferences.

Principle Engraving: In the third stage, we fine-tune the original LLM (the base model) on the self-aligned responses, generated by the LLM itself through prompting, while pruning the principles and demonstrations for the fine-tuned model.

--As large language models are increasingly deployed in sensitive environments, fingerprinting attacks pose significant privacy and security risks. We present a study of LLM fingerprinting from both offensive and defensive perspectives. Our attack methodology uses reinforcement learning to automatically optimize query selection, achieving better fingerprinting accuracy with only 3 queries compared to randomly selecting 3 queries from the same pool. Our defensive approach employs semantic-preserving output filtering through a secondary LLM to obfuscate model identity while maintaining semantic integrity. The defensive method reduces fingerprinting accuracy across tested models while preserving output quality. These contributions show the potential to improve fingerprinting tools capabilities while providing practical mitigation strategies against fingerprinting attacks. Large language models (LLMs) have become ubiquitous across industries, from customer service chatbots to code generation tools and content creation platforms. As organizations increasingly rely on these models for sensitive applications, the ability to identify which specific model generated a given text--known as LLM fingerprinting--has emerged as a critical security concern.