This work evaluates pre-trained embeddings from four state-of-the-art open-source SFMs across three datasets in zero-shot and k-shot cross-lingual settings for multi-class PD severity assessment and shows that pre-trained speech embeddings enable meaningful cross-lingual transfer, although performance is sensitive to d...
Accident-process roles learned from construction narratives remained identifiable in other sectors and an independent organisational setting, and can support assisted coding, expert review, cross-sector analysis, and prevention-oriented analysis of large accident-report collections.
Aho Yapi, P. Latouche, Arnaud Guillin et al.· 0 citations
The Memory Decision Layer (MDL) is proposed, a zero-parameter memory decision controller situated between the retrieval and generation stages that reduces the hallucination rate under conflicting memories by about 56.04% in general scenarios and approaches zero hallucination in high-risk scenarios.
Yi-Ming Zhang, Jing-Hong Zhang, Hao-Ran Zhao et al.· 0 citations
The QuranicMMLU offers a rigorous, linguistically grounded framework for evaluating Arabic NLP in the Quranic domain and benchmark 12 systems on these items find that the Islamic-specialized model leads, yet every system scores higher on multiple-choice accuracy than open-ended answer quality.
R. El Ghali, Umm Kulsoom, Anas Madkoor et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
This work introduces RecreationWorld, a five-platform framework built around recreation: given a running reference, an agent must discover its behavior and build a faithful implementation with no prescribed workflow, and introduces RecreationBench, comprising 250 diverse tasks across domains and platforms.
Shuai Bai, Jia-Yong Deng, Si-Cheng Fan et al.· 1 citation
Moral Entropy is introduced, a Bayesian framework that keeps a full posterior over the true label and decomposes its entropy into aleatoric uncertainty (irreducible disagreement about the moral content) and epistemic uncertainty (from insufficient or noisy annotation) -- and lets any heuristic consensus rule be audited...
TrialAtlas is introduced, a memory-augmented multi-agent research organization for CDP that mirrors this collaborative process by coordinating specialized agents for literature synthesis, competitive trial intelligence, regulatory precedent analysis, and integrated reasoning over trial design and development risk.
Jia-Cheng Lin, Zifeng Wang, Zheng Chen et al.· 0 citations
To improve span localization, ASR encoder states are reuse ASR encoder states as sentence-level representations and fuse them with textual embeddings to let lightweight span locators exploit speech information without running a separate audio encoder.
Steffen Freisinger, Philipp Seeberger, Thomas Ranzenberger et al.· 0 citations
It is shown that on speaker-disjoint, leakage-filtered held-out subsets, the per-aetiology models outperform the mixed baseline across all three target aetiologies: cerebral palsy, Parkinson's disease and amyotrophic lateral sclerosis.
B. Muller, Antonio Armando Ortiz Barrañón, L. Roberts· 0 citations
This work introduces PRISM-BN, a controlled corpus of 5054 BN-grounded descriptions paired with discrete reference BNs containing variables, states, directed edges, root priors, and full multi-parent CPDs across five domains and defines a benchmark with semantic node and state alignment, conditional structural scoring,...
Amartya Bhattacharya, Nikhil Singh, Neeti Pokhriyal et al.· 0 citations
Word Error Rate (WER), the most commonly used metric for Automatic Speech Recognition (ASR), treats every lexical deviation from the reference as equally costly, regardless of whether it changes meaning. This raises the question: does WER actually track how humans judge ASR transcript quality? We introduce HATS-en, an...
Hritika Sharma, Thibault Bañeras-Roux, A. Pinto et al.· 0 citations
CharBM25 is established as an effective, GPU-free retrieval strategy that matches or exceeds dense retrieval at negligible computational cost, and combining it with a general-purpose LLM of 12B+ parameters delivers reliable, training-free Devanagari post-OCR correction without task-specific fine-tuning.
Flint-interactive gives AI agents a language for creating interactive visualizations. Click a chart, filter data, highlight related content, and explore beyond the view the author originally chose. The post Flint-interactive: A language for creating interactive data visualizations appeared first on Microsoft Research.
A new computational approach reveals how cells communicate to assemble into tissues. This could accelerate stem cell engineering, improve disease models, and inform therapies.
Microsoft Research Blog· microsoft.comOct 15, 2026
Zero-knowledge proofs can verify computing results without rerunning workloads or exposing sensitive data. Magnetar helps bring them into the mainstream with support for standard programming languages, major speed gains, and quantum-resistant security. The post Proof-based general-purpose computing with Magnetar appeared first on Microsoft Research.
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
