{"company":{"id":"apple","name":"Apple","homepage":"https://www.apple.com/"},"generated_at":"2026-10-19T11:58:24.907Z","sources":[{"id":"apple-ml:feed","label":"Apple Machine Learning Research blog","kind":"Announcement","url":"https://machinelearning.apple.com/rss.xml","last_checked":"2026-10-18T16:20:44.009Z","last_success":"2026-10-18T16:20:44.009Z","error":null,"state":"ok"}],"social":[],"daily":[{"day":"2026-10-19","company":"apple","material":false,"item_ids":[],"latest_change_at":0,"text":"No material verified update from Apple's official sources on 2026-10-19 (UTC). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-18","company":"apple","material":false,"item_ids":[],"latest_change_at":0,"text":"No material verified update from Apple's official sources on 2026-10-18 (UTC). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-17","company":"apple","material":false,"item_ids":[],"latest_change_at":0,"text":"No material verified update from Apple's official sources on 2026-10-17 (UTC). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-16","company":"apple","material":false,"item_ids":[],"latest_change_at":0,"text":"No material verified update from Apple's official sources on 2026-10-16 (UTC). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-15","company":"apple","material":true,"item_ids":["track:cabebbc6d725875b0784cd21","track:e281ad6449dd1d34841c4242","track:e1e8e56e5379e430ebe65301"],"latest_change_at":1792022400000,"text":"Apple published 3 official updates on 2026-10-15 (UTC): Plan or Retrieve First? A Scale Study of Tool Discovery Architectures (Apple Machine Learning Research blog); Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents (Apple Machine Learning Research blog); Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio (Apple Machine Learning Research blog). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[{"title":"Plan or Retrieve First? A Scale Study of Tool Discovery Architectures","url":"https://machinelearning.apple.com/research/plan-retrieve-tool-discovery","source":"Apple Machine Learning Research blog"},{"title":"Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents","url":"https://machinelearning.apple.com/research/selection-structured-reasoning","source":"Apple Machine Learning Research blog"},{"title":"Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio","url":"https://machinelearning.apple.com/research/mlx-distributed-inference","source":"Apple Machine Learning Research blog"}],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-14","company":"apple","material":true,"item_ids":["track:0a7045ce890ff11a16bbe80e","track:671a5a552d5031a7797cb958"],"latest_change_at":1791936000000,"text":"Apple published 2 official updates on 2026-10-14 (UTC): The Design Space of Tri-Modal Masked Diffusion Models (Apple Machine Learning Research blog); Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models (Apple Machine Learning Research blog). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[{"title":"The Design Space of Tri-Modal Masked Diffusion Models","url":"https://machinelearning.apple.com/research/tri-modal-masked-diffusion","source":"Apple Machine Learning Research blog"},{"title":"Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models","url":"https://machinelearning.apple.com/research/code-switching-detection","source":"Apple Machine Learning Research blog"}],"generated_at":"2026-10-19T11:58:24.907Z"},{"day":"2026-10-13","company":"apple","material":true,"item_ids":["track:ce67544aed8b4d0f51cfe28e"],"latest_change_at":1791849600000,"text":"Apple published 1 official update on 2026-10-13 (UTC): Mix, Don't Tune: Bilingual Pre-Training Outperforms Hyperparameter Search in Data-Constrained Settings (Apple Machine Learning Research blog). Covers Apple's official Apple Machine Learning Research blog. It does not cover social posts or independent reporting.","links":[{"title":"Mix, Don't Tune: Bilingual Pre-Training Outperforms Hyperparameter Search in Data-Constrained Settings","url":"https://machinelearning.apple.com/research/mix-dont-tune","source":"Apple Machine Learning Research blog"}],"generated_at":"2026-10-19T11:58:24.907Z"}],"items":[{"id":"track:cabebbc6d725875b0784cd21","title":"Plan or Retrieve First? A Scale Study of Tool Discovery Architectures","url":"https://machinelearning.apple.com/research/plan-retrieve-tool-discovery","summary":"Tool-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 retrieves similar…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-15T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:e281ad6449dd1d34841c4242","title":"Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents","url":"https://machinelearning.apple.com/research/selection-structured-reasoning","summary":"Multimodal 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, we introduce Selection-based…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-15T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:e1e8e56e5379e430ebe65301","title":"Local and Distributed Large LLM Inference with MLX and RDMA on Mac Studio","url":"https://machinelearning.apple.com/research/mlx-distributed-inference","summary":"MLX on 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…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-15T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:0a7045ce890ff11a16bbe80e","title":"The Design Space of Tri-Modal Masked Diffusion Models","url":"https://machinelearning.apple.com/research/tri-modal-masked-diffusion","summary":"Discrete 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 Models (MDM) pretrained from…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-14T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:671a5a552d5031a7797cb958","title":"Code-Switching Detection via Language-Selective Dimensions in Speech Foundation Models","url":"https://machinelearning.apple.com/research/code-switching-detection","summary":"Code-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 training data. One…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-14T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:ce67544aed8b4d0f51cfe28e","title":"Mix, Don't Tune: Bilingual Pre-Training Outperforms Hyperparameter Search in Data-Constrained Settings","url":"https://machinelearning.apple.com/research/mix-dont-tune","summary":"For 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 data…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-13T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:9941efcb9890c54b6223b036","title":"Self-Compacting Language Model Agents","url":"https://machinelearning.apple.com/research/self-compacting-language-agents","summary":"Long 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 pay no…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-12T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:4e340446a37165a2ba7e323c","title":"Minions: A Dynamic and Distributed Virtual GPU Cloud Built on Idle Apple Silicon Devices","url":"https://machinelearning.apple.com/research/distributed-gpu-cloud","summary":"The 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 are equipped with powerful…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-12T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:5e1213e6b8cbe53e12795c0d","title":"Large Language Models for Machine Translation Quality Annotation: Humans and Models Are Both Challenged","url":"https://machinelearning.apple.com/research/llm-translation-quality-annotation","summary":"Large 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…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-12T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null},{"id":"track:d00eca9ba29392ceb8af1a3d","title":"Normalizing Trajectory Models","url":"https://machinelearning.apple.com/research/normalizing-trajectory-models","summary":"Diffusion-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 objectives, but sacrifice the likelihood…","source":"Apple Machine Learning Research blog","category":"announcement","published_at":"2026-10-08T00:00:00.000Z","first_seen_at":"2026-10-16T19:20:53.168Z","same_as":null}],"guth_coverage":[],"limits":"Official sources listed in \"sources\" only. Social posts and independent reporting are not covered."}