INFUSER is introduced, an iterative co-training framework with two co-evolving roles: a Generator that drafts questions and reference golden answers from a pool of unstructured, automatically collected documents, and a Solver that improves by training on them, and DuGRPO, a dual-normalized variant of GRPO, for generato...
Neural operators have achieved strong performance in learning solution operators of partial differential equations (PDEs), but their inherently continuous representations struggle to capture discontinuities and sharp transitions. Existing approaches typically approximate such features within continuous function spaces,...
Ha Dang, Sebastian Schmidt, Juergen Hesser· 0 citations
This work studies causal retention: whether a frozen learned state answers a mechanism-probe map fixed independently of training, including action, context, direct target, value, and delay, which is distinct from task sufficiency and source-domain decodability.
Sheng-Jun Zhang, Ting Liu, Dong Xie et al.· 0 citations
These results establish persona-set geometry as a task-agnostic mechanism for eliciting divergent LLM outputs, and support persona diversification as a reusable complement to prompt optimization.
Sang-Bin Moon, Nicole Cho, Daniel Borrajo et al.· 0 citations
Reach audiences
Advertise in front of researchers, engineers, and readers.
A Track-and-Stop style algorithm for best-arm identification with prescribed confidence is proposed, and it is proved that the algorithm almost surely achieves the asymptotically optimal cost as the error tends to zero.
The results suggest that the distribution of additional reasoning paths generated by the algorithm is heavy-tailed and that substantial capacity to overcharge remains even under the best possible audit designed to keep the false-positive rate below $\alpha = 0.1$.
Tori Qiu, Ander Artola Velasco, M. Gomez-Rodriguez· 0 citations
Gaussian processes constitute a cornerstone of probabilistic machine learning, yet scaling them to large datasets typically forces a trade-off between computational efficiency and model fidelity. This work bridges this gap by presenting an adaptive multi-resolution Gaussian process framework that is both scalable and e...
Yan-Chuang Cao, Jun Liu, Teng-Chao Yu et al.· 0 citations
Large language models (LLMs) as a judge enable scalable evaluation, but their judgments can be sensitive to response order and, even after removing such position effects, can still diverge systematically from human preferences.We introduce DIAL, a unified framework that combines abundant LLM comparisons with limited hu...
Zesheng Cai, Yingqi Fan, Sichang Chen et al.· 0 citations
CryoEM map interpretation requires features that are spatially localized, consistent across samples, and informative across spatial scales. Most deep learning methods for map annotation extract features from fixed voxel grids. However, implicit neural representations (INRs) are able to model volumetric data as scale-ag...
Phillip Lo, Sudarshan Babu, Dari Kimanius et al.· 0 citations
The raspberry-sesame dual herbal in-situ gel demonstrated favourable physicochemical properties and appreciable in-vitro anti-inflammatory activity, suggesting its potential as a safe topical herbal alternative.
G. Ramalingam, TN Uma Maheshwari, S. Kumar· Journal of Clinical and Diag...· 0 citations
AI-workflow governance cannot be reduced to checking the final answer: an apparently safe answer may rest on a fragile evidence path that ordinary evaluation cannot see, localize or govern, so SSEG moves governance below surface-level output checking into auditable, path-specific decisions about intervention, revalidat...
M. Dixon, Bertrand Nortier, Miquel Noguer I. Alonso· 0 citations
LLM-as-a-Judge is a flexible framework for text evaluation, which allows us to obtain scores for the quality of a given text from various perspectives by changing the prompt template. Two main challenges in using LLM-as-a-Judge are computational cost of inference using a large language model (LLM), especially when eval...
Jennifer Neville did not want to go into computer science—but that’s exactly where she landed. Neville discusses the starts and stops that led to her professional sweet spot and her work identifying “surprising failures” making it hard for AI to handle complexity. The post What AI gets wrong and what failure teaches us appeared first on Microsoft Research.
