3,698 items tracked: 33 reported by two or more outlets, the rest by a single source. Most-corroborated first.
Repeated training on model-generated data can degrade later models.
Biology doesn't operate in silos, and neither should the AI representation of it.
Coding agents can now write, run, and debug programs with little human help.
On-policy distillation (OPD) trains students using teacher feedback on their own sampled responses, yet how prompt choice shapes transfer across teacher-student pairs remains poorly understood.
Historical street-view imagery records urban evolution, but uneven coverage leaves substantial gaps in historical records.
Language-model agents are usually trained by reinforcement learning from one reward per episode, and privileged self-distillation enriches it by letting the same policy, given a skill, teach its skill-free self through t…
Evaluating the physical consistency of generated videos remains a fundamental challenge.
Process reward models (PRMs) have become a key component for LLMs, as their step-level feedback supports both post-training and test-time reasoning.
Video understanding agents acquire evidence through an executable harness that controls what they observe and how they use those observations.
Optical Character Recognition (OCR) is evolving from plain-text transcription toward general visual intelligence, requiring models to recognize, localize, and reason over textual information in complex visual environment…
We study robotic in-context learning (ICL), an emerging paradigm that enables robots to infer and execute tasks from visual demonstrations.
World models, game simulators, and long-take video creation require coherent scene evolution and sustained dynamics over extended durations.
Few-step streaming audio--video generation requires both causal modeling and step distillation, yet standard training recipes face two context-related challenges.
Text-conditioned full-body human-object interaction (HOI) generation requires synthesizing human motion and object trajectories that match the input text while remaining precisely coordinated over time.
Linear attention replaces growing KV caches with fixed-size recurrent states, yet these persistent states can become a substantial memory bottleneck under concurrent serving.
Since launching a year ago, the Microsoft Research Asia — Singapore lab has established a strong foundation, deepened collaboration across government, academia, and industry, and explored how frontier AI research can cre…
Modern AI models are aligned through post-training to adapt them to downstream tasks.
Reinforcement learning from verifiable rewards (RLVR) frequently reuses rollouts across multiple policy updates, increasing the mismatch between the current policy and the data-generating policy.
In response to the growing demand for long sequences in agentic and reasoning use cases, many state-of-the-art LLMs combine multiple variants of attention to mitigate the quadratic complexity of traditional softmax atten…
Mixture-of-experts (MoE) models activate few experts per token, yet batched decoding can access nearly the entire expert pool, making expert-weight traffic a major bottleneck.
Personal Graphical User Interface (GUI) agents rely on interaction history to infer what a user wants from ambiguous instructions and to anticipate recurring routines.
Multimodal large language models can expose task-relevant evidence through generation, but producing useful evidence does not by itself determine how it enters a retrieval embedding.
Evaluating recommender systems before deployment requires simulators that remain realistic over long, closed-loop interaction cycles.
Sparse attention visits few keys, yet finding them can still require scoring every key with every indexer head.
We introduce Adaptive LeWorldModel (ALeWM), a world model based on a joint-embedding predictive architecture (JEPA) that learns to concentrate predictive information in compact prefixes of a wide latent representation.
Reliable progress in offline policy learning depends on careful reporting, well-tuned baselines, and evaluation across diverse conditions.
Optimizing embedding systems for production recommendation pipelines demands systematic exploration that consumes disproportionate engineering effort at scale.
In cross-lingual zero-shot text-to-speech, the accent of the reference leaks into the target speech.
We present TimeBraid, a series of unified time-series and language models that align pretrained language models and pretrained time-series foundation models through interleaved global residual attention layers.
Robots are getting smarter, but how can their hardware match that growth?
Custom-made molecules are advancing medicine, materials, and agriculture, but producing them is slow and expensive.
What if pathology foundation models could do more with less?
Skala 1.1, the updated deep-learning exchange-correlation functional from Microsoft Research, provides greater accuracy, expanded accessibility across the computational chemistry ecosystem, and a living benchmark to trac…