Windows Recall is one of the biggest AI features Microsoft has added to Windows 11. It's useful, but it comes with some privacy trade-offs. If you're a Windows 11 user, and specifically an owner of a Microsoft Copilot+ PC that has the necessary AI processing power, you can now take advantage of Windows Recall. It works like a memory for your computer, using screenshots to keep track of everything you're doing on Windows, and then letting you search back through it. The idea is to give you something like your web browsing history, but for the whole computer.

Reinforcement learning (RL) has emerged as a pivotal technique for fine-tuning large language models (LLMs) on specific tasks. However, prevailing RL fine-tuning methods predominantly rely on PPO and its variants. Though these algorithms are effective in general RL settings, they often exhibit suboptimal performance and vulnerability to distribution collapse when applied to the fine-tuning of LLMs.

The rapid growth of medical knowledge and increasing complexity of clinical practice pose challenges. In this context, large language models ( LLMs) have demonstrated value; however, inherent limitations remain. Retrieval-augmented generation ( RAG) technologies show potential to enhance their clinical applicability. This study reviewed RAG applications in medicine. We found that research primarily relied on publicly available data, with limited application in private data. For retrieval, approaches commonly relied on English-centric embedding models, while LLMs were mostly generic, with limited use of medical-specific LLMs. For evaluation, automated metrics evaluated generation quality and task performance, whereas human evaluation focused on accuracy, completeness, relevance, and fluency, with insufficient attention to bias and safety. RAG applications were concentrated on question answering, report generation, text summarization, and information extraction. Overall, medical RAG remains at an early stage, requiring advances in clinical validation, cross-linguistic adaptation, and support for low-resource settings to enable trustworthy and responsible global use.

Traditional KV cache eviction strategies, which discard less critical KV-pairs based on attention scores, often degrade generation quality, causing context loss or hallucinations. Recent efforts shift toward KV merging, merging eviction tokens with retention tokens based on similarity. However, in multimodal scenarios, distributional biases across modality tokens and attentional biases in cross-modal interactions limit its effectiveness. This work introduces FlowMM, an adaptive framework for cross-modal information flow-guided multimodal KV cache merging. FlowMM leverages cross-modal information flow to dynamically apply layer-specific merging strategies, capturing modality-specific patterns while preserving contextual integrity. Furthermore, we introduce a sensitivity-adaptive token matching mechanism that jointly evaluates token similarity and task-critical sensitivity, merging low-risk tokens while safeguarding high-sensitivity ones. Extensive experiments across diverse leading MLLMs show that FlowMM reduces KV cache memory by 80% to 95% and decoding latency by 1.3-1.8x, while maintaining competitive task performance.

In-context learning (ICL) enables large language models to perform new tasks by conditioning on a sequence of examples. Most prior work reasonably and intuitively assumes that which examples are chosen has a far greater effect on performance than how those examples are ordered, leading to a focus on example selection. We revisit this assumption and conduct a systematic comparison between the effect of selection and ordering. Through controlled experiments on both classification and generation tasks, using multiple open-source model families (0.5B to 27B parameters) and GPT-5, we find that the variance in performance due to different example orderings is comparable to that from using entirely different example sets. Furthermore, we show that strong orderings can be identified using only a development set, achieving performance close to an oracle that selects the best ordering based on test labels. Our findings highlight the equal and intertwined importance of example selection and ordering in prompt design, calling for a reexamination of the assumptions held in ICL.

The Vision-Language-Action models (VLA) have achieved significant advances in robotic manipulation recently. However, vision-only VLA models create fundamental limitations, particularly in perceiving interactive and manipulation dynamic processes. This paper proposes Audio-VLA, a multimodal manipulation policy that leverages contact audio to perceive contact events and dynamic process feedback. Audio-VLA overcomes the vision-only constraints of VLA models. Additionally, this paper introduces the Task Completion Rate (TCR) metric to systematically evaluate dynamic operational processes. Audio-VLA employs pre-trained DINOv2 and SigLIP as visual encoders, AudioCLIP as the audio encoder, and Llama2 as the large language model backbone. We apply LoRA fine-tuning to these pre-trained modules to achieve robust cross-modal understanding of both visual and acoustic inputs. A multimodal projection layer aligns features from different modalities into the same feature space. Moreover RLBench and LIBERO simulation environments are enhanced by adding collision-based audio generation to provide realistic sound feedback during object interactions. Since current robotic manipulation evaluations focus on final outcomes rather than providing systematic assessment of dynamic operational processes, the proposed TCR metric measures how well robots perceive dynamic processes during manipulation, creating a more comprehensive evaluation metric. Extensive experiments on LIBERO, RLBench, and two real-world tasks demonstrate Audio-VLA's superior performance over vision-only comparative methods, while the TCR metric effectively quantifies dynamic process perception capabilities.

Large language models often respond to ambiguous requests by implicitly committing to one interpretation. Intent misunderstandings can frustrate users and create safety risks. To address this, we propose generating multiple interpretation-answer pairs in a single structured response to ambiguous requests. Our models are trained with reinforcement learning and customized reward functions using multiple valid answers as supervision. Experiments on conversational question answering and semantic parsing demonstrate that our method achieves higher coverage of valid answers than baseline approaches. Human evaluation confirms that predicted interpretations are highly aligned with their answers. Our approach promotes transparency with explicit interpretations, achieves efficiency by requiring only one generation step, and supports downstream applications through its structured output format.

The rapid expansion of asynchronous remote care has intensified provider workload, creating demand for AI systems that can assist clinicians in managing patient queries more efficiently. The MEDIQA-WV 2025 shared task addresses this challenge by focusing on generating free-text responses to wound care queries paired with images. In this work, we present two complementary approaches developed for the English track. The first leverages a mined prompting strategy, where training data is embedded and the top-k most similar examples are retrieved to serve as few-shot demonstrations during generation. The second approach builds on a metadata ablation study, which identified four metadata attributes that consistently enhance response quality. We train classifiers to predict these attributes for test cases and incorporate them into the generation pipeline, dynamically adjusting outputs based on prediction confidence. Experimental results demonstrate that mined prompting improves response relevance, while metadata-guided generation further refines clinical precision. Together, these methods highlight promising directions for developing AI-driven tools that can provide reliable and efficient wound care support.

Large Language Models (LLMs) have demonstrated impressive capabilities in solving complex tasks, including those requiring a certain level of reasoning. In this paper, we focus on state tracking, a problem where models need to keep track of the state governing a number of entities. To isolate the state tracking component from other factors, we propose a benchmark based on three well-defined state tracking tasks and analyse the performance of LLMs in different scenarios. The results indicate that the recent generation of LLMs (specifically, GPT-4 and Llama3) are capable of tracking state, especially when integrated with mechanisms such as Chain of Thought. However, models from the former generation, while understanding the task and being able to solve it at the initial stages, often fail at this task after a certain number of steps.

Frontier LLMs are optimised around high-resource assumptions about language, knowledge, devices, and connectivity. Whilst widely accessible, they often misfit conditions in the Global South. As a result, users must often perform additional work to make these systems usable. We term this alignment debt: the user-side burden that arises when AI systems fail to align with cultural, linguistic, infrastructural, or epistemic contexts. We develop and validate a four-part taxonomy of alignment debt through a survey of 411 AI users in Kenya and Nigeria. Among respondents measurable on this taxonomy (n = 385), prevalence is: Cultural and Linguistic (51.9%), Infrastructural (43.1%), Epistemic (33.8%), and Interaction (14.0%). Country comparisons show a divergence in Infrastructural and Interaction debt, challenging one-size-fits-Africa assumptions. Alignment debt is associated with compensatory labour, but responses vary by debt type: users facing Epistemic challenges verify outputs at significantly higher rates (91.5% vs. 80.8%; p = 0.037), and verification intensity correlates with cumulative debt burden (Spearmans rho = 0.147, p = 0.004). In contrast, Infrastructural and Interaction debts show weak or null associations with verification, indicating that some forms of misalignment cannot be resolved through verification alone. These findings show that fairness must be judged not only by model metrics but also by the burden imposed on users at the margins, compelling context-aware safeguards that alleviate alignment debt in Global South settings. The alignment debt framework provides an empirically grounded way to measure user burden, informing both design practice and emerging African AI governance efforts.