Apple (United States)

An Apple team aligned wrist photoplethysmography (PPG, a green-light measure of microvascular blood-volume change) with meal descriptions to train the Nutrition PPG Language Model (NPLM) and test whether passively collected pulse waveforms carry dietary information. On 1,122,834 meal descriptions from 19,340 participants, near-meal PPG raised the likelihood of the correct meal text to a win rate of 0.69, against 0.60 for a control that kept the same participant and time of day but broke the meal pairing, and 0.50 for across-subject shuffling. Without retraining, the signal raised postprandial satiety prediction AUC from 0.64 to 0.71 in an independent cohort, which the authors frame as augmenting text-based logging rather than replacing it.

An Apple team aligned wrist photoplethysmography (PPG, a green-light measure of microvascular blood-volume change) with meal descriptions to train the Nutrition PPG Language Model (NPLM) and test whether passively collected pulse waveforms carry dietary information. On 1,122,834 meal descriptions from 19,340 participants, near-meal PPG raised the likelihood of the correct meal text to a win rate of 0.69, against 0.60 for a control that kept the same participant and time of day but broke the meal pairing, and 0.50 for across-subject shuffling. Without retraining, the signal raised postprandial satiety prediction AUC from 0.64 to 0.71 in an independent cohort, which the authors frame as augmenting text-based logging rather than replacing it.

For a fixed feature distribution and an arbitrary model class, agnostic learning (training a discriminator that matches the best member of the class) and multiaccuracy (training a generator whose labels no observer in the class can distinguish from real ones) are proved equivalent in statistical and computational complexity. Only the direction from agnostic learning to multiaccuracy was known; the converse was blocked because a single multiaccurate predictor can fail to beat random guessing even after calibration. The new reduction repeatedly queries a multiaccuracy oracle, splits by prediction level sets and averages the iterates, yielding a predictor that is both multiaccurate and calibrated and thus performs agnostic learning.

For a fixed feature distribution and an arbitrary model class, agnostic learning (training a discriminator that matches the best member of the class) and multiaccuracy (training a generator whose labels no observer in the class can distinguish from real ones) are proved equivalent in statistical and computational complexity. Only the direction from agnostic learning to multiaccuracy was known; the converse was blocked because a single multiaccurate predictor can fail to beat random guessing even after calibration. The new reduction repeatedly queries a multiaccuracy oracle, splits by prediction level sets and averages the iterates, yielding a predictor that is both multiaccurate and calibrated and thus performs agnostic learning.

The authors introduce MS-ECG-FM, an ECG foundation model pre-trained by contrastive alignment to several clinical note types — ECG machine reports, cardiologist ECG reports, echocardiography, chest X-ray and discharge summaries; on linear-probe evaluation over 98 labels across multiple datasets it reaches a macro-averaged AUROC of 93.6 versus 91.5 for ECGFounder and 91.2 for MELP, with clear gains on structural heart disease and maintained leadership in single-lead and reduced-lead configurations.

The authors introduce MS-ECG-FM, an ECG foundation model pre-trained by contrastive alignment to several clinical note types — ECG machine reports, cardiologist ECG reports, echocardiography, chest X-ray and discharge summaries; on linear-probe evaluation over 98 labels across multiple datasets it reaches a macro-averaged AUROC of 93.6 versus 91.5 for ECGFounder and 91.2 for MELP, with clear gains on structural heart disease and maintained leadership in single-lead and reduced-lead configurations.

The study operationalizes homogenization in multi-agent systems through conformity, polarization, and inertia, evaluates it on code generation, hiring, and peer review, and finds that agent interactions reduce diversity and amplify shared failures, while sampling stochasticity and mixed-model configurations fail to mitigate these risks.

The study operationalizes homogenization in multi-agent systems through conformity, polarization, and inertia, evaluates it on code generation, hiring, and peer review, and finds that agent interactions reduce diversity and amplify shared failures, while sampling stochasticity and mixed-model configurations fail to mitigate these risks.

DeReAct introduces a modular agent architecture that separates action authorization and completion control from a single LLM policy, delegating them to a Critic that validates proposed actions before execution and a Context Manager that reconstructs an environment-supported State and certifies task completion; across GAIA and SWE-bench Verified it improves Pass@1 most for weaker Brain models (6.5–7.0 points for Qwen3-Coder-480B and 4.2–5.2 points for Claude Sonnet 4.5), with gains diminishing as Brain capability increases, while on Claude Opus 4.5 Pass@1 remains comparable to ReAct but trajectories are more evidence-complete and constraint-satisfying.

DeReAct introduces a modular agent architecture that separates action authorization and completion control from a single LLM policy, delegating them to a Critic that validates proposed actions before execution and a Context Manager that reconstructs an environment-supported State and certifies task completion; across GAIA and SWE-bench Verified it improves Pass@1 most for weaker Brain models (6.5–7.0 points for Qwen3-Coder-480B and 4.2–5.2 points for Claude Sonnet 4.5), with gains diminishing as Brain capability increases, while on Claude Opus 4.5 Pass@1 remains comparable to ReAct but trajectories are more evidence-complete and constraint-satisfying.

Addressing hallucinations in Multimodal Large Language Models (MLLMs), this work proposes Causal-Invariant Masking (CIM), which measures the semantic shift between original predictions and those conditioned on a causally-focused view to decompose uncertainty types, introduces Semantic Divergence as the core UQ metric with theoretical evidence that it converges to the variance of the model's sensitivity to non-causal correlations, and further proposes Expected Embedding Drift (EED), a fast geometric proxy metric; experiments report state-of-the-art performance on various benchmarks, with EED accelerating by nearly 50% at comparable performance.

Addressing hallucinations in Multimodal Large Language Models (MLLMs), this work proposes Causal-Invariant Masking (CIM), which measures the semantic shift between original predictions and those conditioned on a causally-focused view to decompose uncertainty types, introduces Semantic Divergence as the core UQ metric with theoretical evidence that it converges to the variance of the model's sensitivity to non-causal correlations, and further proposes Expected Embedding Drift (EED), a fast geometric proxy metric; experiments report state-of-the-art performance on various benchmarks, with EED accelerating by nearly 50% at comparable performance.