There is growing concern that AI chatbots might fuel delusional beliefs in users. Some have suggested that humans and chatbots mutually reinforce false beliefs over time, but quantitative evidence is lacking. Using a unique dataset of chat logs from individuals who exhibited delusional thinking, we developed a latent s...
Ashish Mehta, Jared Moore, Jacy Reese Anthis et al.· 0 citations
This study investigates the structural organisation of Dante's Divina Commedia through a symbolic representation based on vowel-consonant (V/C) encoding. Modelling the resulting sequence through Markov chains provides parsimonious measures of local graphemic dependence, capturing persistence and alternation patterns. A...
Personal intelligence requires strong capabilities under strict privacy and efficiency constraints. Unlike cloud-first intelligence, which relies on powerful but costly and privacy-vulnerable large language models (LLMs), we focus on local-first intelligence, emphasizing on-device small language models (SLMs), which en...
Hang Zeng, Xiangyu Liu, Yong Hu et al.· 0 citations
MedPRMBench is introduced, a step-level medical reasoning benchmark with 14 clinical error types, including five new and two clinically re-scoped types, and rule-defined severity grades, and the training signal replicates across model families.
Despite advances in multilingual capabilities, most large language models (LLMs) remain English-centric in their training and, crucially, in the reasoning traces they produce: even for non-English problems, they predominantly reason in English, leaving non-English users with intermediate reasoning they cannot follow. W...
Daniil Gurgurov, Tom R\"ohr, Sebastian von Rohrscheidt et al.· 0 citations
Human perception interprets video dynamics by constructing structured mental representations of entities, actions, and temporal boundaries before engaging in high-level reasoning. In contrast, current Video-LLMs typically rely on unstructured Chain-of-Thought (CoT), where crucial visual cues are obscured by verbose nar...
Zinuo Li, Yongxin Guo, Jun Liu et al.· 0 citations
Reward models are crucial for aligning large language models (LLMs) with human values and intentions. Existing approaches follow either Generative (GRMs) or Discriminative (DRMs) paradigms, yet both suffer from limitations: GRMs typically demand costly point-wise supervision, while DRMs produce uncalibrated relative sc...
Reinforcement Learning (RL) has emerged as a key driver for LLM reasoning. This capability is equally pivotal in long-context scenarios---such as long-dialogue understanding and structured data analysis---where the challenge extends beyond consuming tokens to performing rigorous deduction. While existing efforts focus...
Jailbreak attacks on safety-aligned LLMs are increasingly effective, yet attack success alone does not reveal whether compliance is driven by preserved task intent, suppressed safety signals, or attack-specific artifacts. We use Pearl's front-door adjustment to motivate separating task-relevant semantics from safety-re...
Yao Zhou, Zeen Song, Wenwen Qiang et al.· 0 citations
AfriEconQA, a benchmark for document-grounded question answering built from 204 World Bank economic reports on African economies, is introduced, a corpus whose claims are tied to specific fiscal periods, countries, and projection states.
E. Ajayi, Mustapha Alaba, D. Stephen et al.· 0 citations
As Natural Language Generation (NLG) dominates modern NLP, scalable evaluation remains a critical bottleneck. Consequently, LLM-as-a-judge (LaaJ) adoption has accelerated rapidly, appearing in more papers than human evaluation in 2025. This pivotal shift motivates a critical analysis of current evaluation practices. Ov...
Jing Yang, Nils Feldhus, Salar Mohtaj et al.· 0 citations
Understanding the alignment between large language models (LLMs) and human brain activity can reveal computational principles underlying language processing. We introduce a fine-grained input attribution method to identify the specific words most important for brain-LLM alignment, and leverage it to study a contentious...
Michela Proietti, Roberto Capobianco, Mariya Toneva· 0 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.
