Muse is separate from the existing Meta AI app with its own iOS and Android apps and website. You don't need a Facebook or Instagram account, but you will need a Meta account. On web, all I had to do to log in was enter my phone number and Muse AI already had my relevant accounts linked. After logging in, one of the first things Muse will ask is for you to rename it. I don't know why, but Meta really wants you to name your agent.

Muse, Meta's newly-released AI agent, is now using social media handles once controlled by Muse, the English rock band. The exact circumstances surrounding how the accounts changed hands are unclear, but it has once again raised questions about Meta and other large platforms' ability to commandeer usernames when it suits them. Muse, the band, which trademarked its name in 1999, according to its wiki, has used the @muse moniker on Instagram and on X for years. But earlier this summer fans noticed the band's Instagram username seemed to change to @museband. The change happened in July, according to The Independent, though Reddit users seem to have spotted a change in early June.

Designed to compete with OpenClaw and Instinct, the company says Muse can do everything from sell your car to book you a plane ticket. Meta announced Tuesday the release of Muse, a personal AI agent that people can message to automate digital tasks in a secure cloud environment, all while relying on security and privacy that the company says is "built into it" from the start. Meta says Muse is rolling out today for iOS and Android users in a dedicated Muse app, as well as the website Muse.ai. The company also says users can directly message Muse in WhatsApp to interact with the agent. Meta says users of its AI glasses will soon be able to interact with its Muse agent as well.

Meta has revealed its long-promised personal AI agent that requires "no technical experience" to set up. Called Muse, Meta says the agent is able to complete a wide range of actions on users' behalf, including online shopping, drafting and sending emails, planning trips and other productivity-focused tasks. CEO Mark Zuckerberg has previously said that he wants Meta to create agents that can work proactively to help people get things done, but in a more accessible format than other platforms like OpenClaw. Muse, which has its own dedicated apps and website, is more capable than the Meta AI assistant because it can handle complex tasks and work independently. "Once a person shares a goal with Muse, it helps them develop a personalized plan and coordinate their time and resources, then advances the work on its own," the company explains in a blog post.

Lifelong user interest modeling is crucial for industrial recommender systems, yet existing approaches rely predominantly on ID-based features, suffering from poor generalization on long-tail items and limited semantic expressiveness. While recent work explores multimodal representations for behavior retrieval in the General Search Unit (GSU), they often neglect multimodal integration in the fine-grained modeling stage -- the Exact Search Unit (ESU). In this work, we present a systematic analysis of how to effectively leverage multimodal signals across both stages of the two-stage lifelong modeling framework. Our key insight is that simplicity suffices in the GSU: lightweight cosine similarity with high-quality multimodal embeddings outperforms complex retrieval mechanisms. In contrast, the ESU demands richer multimodal sequence modeling and effective ID-multimodal fusion to unlock its full potential. Guided by these principles, we propose MUSE, a simple yet effective multimodal search-based framework. MUSE has been deployed in Taobao display advertising system, enabling 100K-length user behavior sequence modeling and delivering significant gains in top-line metrics with negligible online latency overhead. To foster community research, we share industrial deployment practices and open-source the first large-scale dataset featuring ultra-long behavior sequences paired with high-quality multimodal embeddings. Our code and data is available at https://taobao-mm.github.io.

Achieving a balance between lightweight design and high performance remains a significant challenge for speech enhancement (SE) tasks on resource-constrained devices. Existing state-of-the-art methods, such as MUSE, have established a strong baseline with only 0.51M parameters by introducing a Multi-path Enhanced Taylor (MET) transformer and Deformable Embedding (DE). However, an in-depth analysis reveals that MUSE still suffers from efficiency bottlenecks: the MET module relies on a complex "approximate-compensate" mechanism to mitigate the limitations of Taylor-expansion-based attention, while the offset calculation for deformable embedding introduces additional computational burden. This paper proposes IMSE, a systematically optimized and ultra-lightweight network. We introduce two core innovations: 1) Replacing the MET module with Amplitude-Aware Linear Attention (MALA). MALA fundamentally rectifies the "amplitude-ignoring" problem in linear attention by explicitly preserving the norm information of query vectors in the attention calculation, achieving efficient global modeling without an auxiliary compensation branch. 2) Replacing the DE module with Inception Depthwise Convolution (IDConv). IDConv borrows the Inception concept, decomposing large-kernel operations into efficient parallel branches (square, horizontal, and vertical strips), thereby capturing spectrogram features with extremely low parameter redundancy. Extensive experiments on the VoiceBank+DEMAND dataset demonstrate that, compared to the MUSE baseline, IMSE significantly reduces the parameter count by 16.8\% (from 0.513M to 0.427M) while achieving competitive performance comparable to the state-of-the-art on the PESQ metric (3.373). This study sets a new benchmark for the trade-off between model size and speech quality in ultra-lightweight speech enhancement.

Interfaces for contemporary large language, generative media, and perception AI models are often engineered for single user interaction. We investigate ritual as a design scaffold for developing collaborative, multi-user human-AI engagement. We consider the specific case of an immersive staging of the musical Xanadu performed at UCLA in Spring 2025. During a two-week run, over five hundred audience members contributed sketches and jazzercise moves that vision language models translated to virtual scenery elements and from choreographic prompts. This paper discusses four facets of interaction-as-ritual within the show: audience input as offerings that AI transforms into components of the ritual; performers as ritual guides, demonstrating how to interact with technology and sorting audience members into cohorts; AI systems as instruments "played" by the humans, in which sensing, generative components, and stagecraft create systems that can be mastered over time; and reciprocity of interaction, in which the show's AI machinery guides human behavior as well as being guided by humans, completing a human-AI feedback loop that visibly reshapes the virtual world. Ritual served as a frame for integrating linear narrative, character identity, music and interaction. The production explored how AI systems can support group creativity and play, addressing a critical gap in prevailing single user AI design paradigms.

In this work, we introduce MUSE (Model-based Uncertainty-aware Similarity Estimation), a training-free framework designed for model-based zero-shot 2D object detection and segmentation. MUSE leverages 2D multi-view templates rendered from 3D unseen objects and 2D object proposals extracted from input query images. In the embedding stage, it integrates class and patch embeddings, where the patch embeddings are normalized using generalized mean pooling (GeM) to capture both global and local representations efficiently. During the matching stage, MUSE employs a joint similarity metric that combines absolute and relative similarity scores, enhancing the robustness of matching under challenging scenarios. Finally, the similarity score is refined through an uncertainty-aware object prior that adjusts for proposal reliability. Without any additional training or fine-tuning, MUSE achieves state-of-the-art performance on the BOP Challenge 2025, ranking first across the Classic Core, H3, and Industrial tracks. These results demonstrate that MUSE offers a powerful and generalizable framework for zero-shot 2D object detection and segmentation.

Large Language Models have demonstrated remarkable capabilities across diverse domains, yet significant challenges persist when deploying them as AI agents for real-world long-horizon tasks. Existing LLM agents suffer from a critical limitation: they are test-time static and cannot learn from experience, lacking the ability to accumulate knowledge and continuously improve on the job. To address this challenge, we propose MUSE, a novel agent framework that introduces an experience-driven, self-evolving system centered around a hierarchical Memory Module. MUSE organizes diverse levels of experience and leverages them to plan and execute long-horizon tasks across multiple applications. After each sub-task execution, the agent autonomously reflects on its trajectory, converting the raw trajectory into structured experience and integrating it back into the Memory Module. This mechanism enables the agent to evolve beyond its static pretrained parameters, fostering continuous learning and self-evolution. We evaluate MUSE on the long-horizon productivity benchmark TAC. It achieves new SOTA performance by a significant margin using only a lightweight Gemini-2.5 Flash model. Sufficient Experiments demonstrate that as the agent autonomously accumulates experience, it exhibits increasingly superior task completion capabilities, as well as robust continuous learning and self-evolution capabilities. Moreover, the accumulated experience from MUSE exhibits strong generalization properties, enabling zero-shot improvement on new tasks. MUSE establishes a new paradigm for AI agents capable of real-world productivity task automation.

We present Multipole Semantic Attention (MuSe), an efficient approximation of softmax attention that combines semantic clustering with multipole expansions from computational physics. Our method addresses the quadratic computational complexity of transformers in the context length by clustering queries and keys separately in their learned representation spaces, enabling a hierarchical two-stage attention mechanism. Unlike prior clustering approaches that group only keys or use unified clustering, we maintain separate clusterings that respect attention's asymmetric treatment of these spaces. We augment centroid-based (monopole) approximations with dipole corrections that capture directional variance within clusters, preserving richer information during training. The method operates as a drop-in replacement for standard attention, requiring only hyperparameter specification without architectural modifications. Our approach achieves $\mathcal{O}(NCD)$ complexity for acausal attention with $C$ clusters and $\mathcal{O}(NCD \log N)$ for causal attention. On isolated attention layers, we demonstrate $3\times$ speedup over CUDNN Flash Attention at 8k context length, with relative squared errors below 20%. For causal attention, we develop a hierarchical block decomposition that combines exact local computation with efficient long-range approximation. In end-to-end pretraining of a 30M parameter model on book-length texts with 16k context, we achieve 12.2% runtime reduction with only 0.36% loss degradation, establishing the viability of multipole approximations for efficient transformer pretraining.