AI News: Apple Machine Learning Research https://ai-news.ypetit.net/sources/apple-ml-research/ Latest publications from Apple Machine Learning Research, aggregated by AI News from the official source. en Fri, 23 Oct 2026 19:08:30 GMT Apple Machine Learning Research: Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studiohttps://machinelearning.apple.com/research/mlx-distributed-inferencemlx-distributed-inferenceThu, 15 Oct 2026 00:00:00 GMTresearchon Mac with Apple silicon lets AI researchers and developers experiment locally with the latest models and methods in a private environment. The open-source array framework leverages the unified memory architecture of Apple silicon, enabling anyone to run new inference and fin… Apple Machine Learning Research: Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agentshttps://machinelearning.apple.com/research/selection-structured-reasoningselection-structured-reasoningThu, 15 Oct 2026 00:00:00 GMTresearchMultimodal agents commonly generate freeform reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substantial inference cost. To address this challenge,… Apple Machine Learning Research: Plan or Retrieve First? A Scale Study of Tool Discovery Architectureshttps://machinelearning.apple.com/research/plan-retrieve-tool-discoveryplan-retrieve-tool-discoveryThu, 15 Oct 2026 00:00:00 GMTresearchTool-augmented LLM agents discover relevant APIs from large registries before planning execution, yet finite context windows prohibit injecting all tool definitions into a single prompt. The dominant approach, which we term Retrieve-then-Plan (RtP), embeds the user query and r… Apple Machine Learning Research: Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Modelshttps://machinelearning.apple.com/research/code-switching-detectioncode-switching-detectionWed, 14 Oct 2026 00:00:00 GMTresearchCode-switching (CS), the alternation between languages within an utterance, is common in multilingual speech but underrepresented in pretraining corpora. Increasing CS coverage requires finding relevant utterances in large speech collections so that they can be included in tra… Apple Machine Learning Research: The Design Space of Tri-Modal Masked Diffusion Modelshttps://machinelearning.apple.com/research/tri-modal-masked-diffusiontri-modal-masked-diffusionWed, 14 Oct 2026 00:00:00 GMTresearchDiscrete diffusion models have emerged as strong alternatives to autoregressive language models, with recent work initializing and finetuning a base unimodal model for bi-modal generation. Diverging from previous approaches, we introduce the first tri-modal Masked Diffusion Mo… Apple Machine Learning Research: Mix, Don't Tune: Bilingual Pre-Training Outperforms Hyperparameter Search in Data-Constrained Settingshttps://machinelearning.apple.com/research/mix-dont-tunemix-dont-tuneTue, 13 Oct 2026 00:00:00 GMTresearchFor most languages of the world, language model pre-training operates in a data-constrained regime where models must repeat their training data many times, degrading generalization. Two remedies exist: aggressive hyperparameter tuning such as high weight decay, and mixing in d… Apple Machine Learning Research: Large Language Models for Machine Translation Quality Annotation: Humans and Models Are Both Challengedhttps://machinelearning.apple.com/research/llm-translation-quality-annotationllm-translation-quality-annotationMon, 12 Oct 2026 00:00:00 GMTresearchLarge Language Models (LLMs) are considered to be a more efficient and cost-effective alternative to human judgment for Machine Translation (MT) evaluation. With MT evaluation spanning a large number of language pairs, domains and levels of annotation granularity, LLMs must be… Apple Machine Learning Research: Minions: A Dynamic and Distributed Virtual GPU Cloud Built on Idle Apple Silicon Deviceshttps://machinelearning.apple.com/research/distributed-gpu-clouddistributed-gpu-cloudMon, 12 Oct 2026 00:00:00 GMTresearchThe increasing demand for GPU-accelerated machine learning inference has exposed limitations in centralized GPU infrastructures, including high cost, limited elasticity, and long wait times under resource constraints. At the same time, modern Apple Silicon–based macOS devices … Apple Machine Learning Research: Self-Compacting Language Model Agentshttps://machinelearning.apple.com/research/self-compacting-language-agentsself-compacting-language-agentsMon, 12 Oct 2026 00:00:00 GMTresearchLong agent traces composed of chains of thought and tool calls accumulate stale content that anchor subsequent generations, and eventually outgrow the context window. Existing scaffolds mitigate it with fixed-interval compaction triggered at a token threshold. Such triggers pa… Apple Machine Learning Research: Normalizing Trajectory Modelshttps://machinelearning.apple.com/research/normalizing-trajectory-modelsnormalizing-trajectory-modelsThu, 08 Oct 2026 00:00:00 GMTresearchDiffusion-based models decompose sampling into many small Gaussian denoising steps, an assumption that breaks down when generation is compressed to a few coarse transitions. Existing few-step methods address this through distillation, consistency training, or adversarial objec…