Accuracy on medical examinations is common evidence offered for the clinical use of language models, yet a health system that intends to run an open model on its own hardware needs to know considerably more: whether the model recognizes when it is likely to be wrong, whether medical fine-tuning improves this ability, w...
Miguel Fernandez-de-Retana, Unai Zulaika, Ruben Sanchez-Corcuera et al.· 0 citations
Looped and depth-grown language models reuse computations across depth and have attracted interest for their reasoning capabilities. Building on prior suggestive evidence of improved in-context-learning (ICL), we investigate how such reuse favors ICL, including how it shapes the acquisition, retention, and mechanisms o...
Ferdinand Kapl, Kaitlin Maile, Stefan Bauer et al.· 0 citations
Continual learning requires a model to keep acquiring new skills and knowledge without compromising its existing capabilities. Parametric methods suffer from catastrophic forgetting and loss of plasticity. Non-parametric methods, such as prompts, skills, and harnesses, leverage in-context learning but are limited by co...
The results show that source deletion should be evaluated behaviorally at the level of the agent and its memory system, not only as a storage operation.
Ehsan Saleh, Roozbeh Nahavandi, Sama Salarian et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
Metadata-conditioned generation is often framed as a data-to-text task in which additional input fields should improve grounding. Museum description generation challenges this assumption because catalogue records are heterogeneous, partially visual, and unevenly documented across geographic regions. We study how metada...
LLM-as-a-Judge evaluation is increasingly applied to multilingual tasks, yet prompting strategies such as Chain-of-Thought and Role Conditioning are recommended based solely on English benchmarks. Whether their effectiveness transfers across judge models and languages remains unknown: prompt-sensitivity studies are lar...
A. Mehrabi, Mohammad Aliannejadi, Clemencia Siro· 0 citations
Language models are compared on question-answering benchmarks that pose each question in one wording, and a reworded question can change which of two models scores higher. We ask whether giving the models tools makes such comparisons more stable. An accuracy gain cannot answer this, because tools change several steps b...
Mateusz Modrzejewski, Tomasz Radzikowski· 0 citations
Long-context models often use answer-bearing evidence less effectively when it appears away from a prompt boundary. We compare this position bias across Transformer, state-space, and hybrid Mamba architectures by moving one gold document through a fixed set of nine distractors. The evaluated Transformers show a beginni...
Aman Behera, Namit Solanki, Mehul Anand· 0 citations
Our previous analysis found that a model's native hashed N-gram table is where continual updates disturb the base least, but its writes needed backpropagation and changed the pretrained table. We present Delta N-gram, a forward-only write path into that table. The pretrained table T^0 is never modified. Evidence is wri...
Single-document LoRA adapters have been proposed as units of knowledge to be merged and retrieved; yet open-book scoring, with the document in the prompt, can credit a document's own adapter (its specialist) for reading that document as readily as for holding it. Training one adapter for each of 150 Wikipedia articles...
Martino M. L. Pulici, Cuong Xuan Chu, Evgeny Kharlamov et al.· 0 citations
Many language-model pipelines end in a closed-set decision, such as selecting a tool. Prompting a large language model for it is slow and order-sensitive. We ask how small an encoder for such decisions can be and whether it should read the options jointly or separately. In a three-seed study at 32 million parameters, p...
Six new safety datasets of sufficient size and difficulty are constructed and it is found that probe error follows two simple scaling laws, which allow safety researchers to determine how to create the best probe for any given model and label budget.
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.
