Law changes over time. A court decides a case under the law in force on that day, and legal QA benchmarks grade answers against today's law. We checked 156,650 decisions of four Thai courts. Up to 67% of them cite law that has since been amended or repealed. We introduce NitiBench+, the sequel of our NitiBench, a Thai...
Pawitsapak Akarajaradwong, S. Nutanong, Chompakorn Chaksangchaichot· 0 citations
As streaming generation becomes the dominant serving paradigm for both text-only and multimodal large language models, ensuring safety throughout the generation process has become a critical challenge. We introduce PGuard, a native, streaming, and plug-in guardrail that transforms safety monitoring from an external ser...
Junyu Lin, Mengjie Wu, Yonghao Dong et al.· 0 citations
Dialectal Arabic speech recognition remains constrained by the scarcity of dialect-resolved transcribed speech despite the abundance of web audio. We introduce GigaSpeech 3, a corpus of 39,000 hours and 10 million weakly transcribed speech segments covering Algerian, Egyptian, Iraqi, Saudi, and Syrian Arabic. Starting...
Junzhe Liu, Yujie Tu, Yifan Yang et al.· 0 citations
We compare how often two released language models give outdated numeric values from Norwegian statutes and regulations: NbAiLab/borealis2-26b-a4b-preview (Borealis 2), which the National Library of Norway further trained from google/gemma-4-26B-A4B-it (Gemma 4), and Gemma 4 itself. The study was preregistered (doi:10.5...
Eirik Botten Nicolaysen· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A systematic comparison across five low-resource African languages and Swahili as a mid-resource contrast is presented, fine-tuning three models from distinct architectures: Phi-4-multimodal-instruct, Whisper-large-v3-turbo, and MMS-1B-all.
Mercy Muchai, Kevin Chege, Nick Mumero Mwangi et al.· 0 citations
Quality-Aware Decoding (QAD) improves machine translation through test-time scaling, generating multiple candidate translations and selecting the final output with a quality-aware decoder. Usually the same candidate budget N is spent on every source sentence, assuming that all inputs require the same amount of search....
Javier Garcia Gilabert, Rico Sennrich, Andr\'e F. T. Martins· 0 citations
Omni-modal large language models (LLMs) have achieved remarkable progress in audio and video understanding. However, existing models struggle with long-form, multi-speaker videos, especially in understanding the entire video and recognizing who says what and when. To characterize the capability limit of current omni LL...
Yuan Ge, Pei Zhang, Xize Cheng et al.· 0 citations
To prepare for future outbreaks, researchers need to know what studies have been done and what data they use. Mapping a large literature by hand is slow, while large language models alone can misstate or conflate sources. We present EpiLens, a lightweight, open-source retrieval-augmented generation package that applies...
Responding to observable behavior is a more robust basis for social adaptation than conditioning on psychological labels and is suggested to be preferred over Oracle overall.
Maryam Shoaeinaeini, Brent Harrison, A. Siddique· arXiv.org· 0 citations
High-performance GPU kernels are essential for efficient LLM deployment, yet optimizing them remains expertise-intensive. Recent LLM-based code generation makes automatic GPU operator generation promising, but operator optimization remains a hardware-aware search problem. Existing LLM-based methods face a granularity m...
Yining Zhang, Mingyang Yi, Chen Wang et al.· 0 citations
alysing how an agent learns with the environment it learns from is a promising direction for self-evolving tool-use agents, and SEAL, a framework for synergistic co-evolution of agents and learning environments is introduced.
Yihao Hu, Zhi-Hao Wen, Xiu-Jin Liu et al.· arXiv.org· 14 citations
ProofRank is introduced, a benchmark curated from challenging competitions that measures several scalable proxies of proof quality and captures important aspects of mathematical proofs beyond correctness, and indicates the need for models that produce proofs which are not only correct but also valuable.
Ivo Petrov, Jasper Dekoninck, Dimitar I. Dimitrov et al.· arXiv.org· 3 citations
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.
