Category
343 papers
This work introduces a general steering technique called Semantic Overlays: small learned adapters applied at chosen prefill positions to a frozen model's residual stream that defends against the broad class of prompt injections that add instructions in untrusted context.
This work introduces Mixture of Channel Experts (MoCE), a structured sparse channel-mixing layer, inspired by MoE, that replaces pointwise (1x1) channel-reduction projections and matches or exceeds dense baselines and prior channel-selection methods while reducing MACs by 16.7% and end-to-end latency.
This work presents MARS (Multi-Agent Relay of Specialized LLMs), a prompt-only framework in which each agent is a topic specialist---dynamic programming, graphs, strings, geometry, and so on---grounded by retrieval-augmented generation over an algorithm-theory corpus.
A reproducible reliability audit of the developer-accessible on-device foundation model is presented, framed as an oversight question: can a user or a resource-constrained developer tell when the model is wrong?
Reach audiences
Advertise in front of researchers, engineers, and readers.
Stochastic analysis sharpens rather than erodes the thesis: the ignition boundary acquires a predicted width, and noise punishes the reactive policy that parks the system against it.
Results show that a small box-level module can reconcile question understanding with precise localization without retraining either backbone, and introduce RefineRank, which closes this gap at the candidate-box level.
Discrete diffusion provides an effective alternative to autoregressive radiology report generation by enabling iterative, bidirectional report refinement.
Whether off-the-shelf Large Language Models (LLMs) can effectively reason about taint flows in Android apps is investigated, and preliminary findings suggest that LLM reasoning may effectively complement traditional static taint analysis.
Results indicate that current LLMs provide uneven safety assurance across Urdu's script varieties, with smaller open-weight models showing substantially higher instability and missed-harm rates than frontier closed models.
This work analyzes a complete corpus of 10,211 inbound scam and spam calls collected over 54 days by an AI voice-agent honeypot that answered callers and kept them talking, and introduced in a companion data descriptor.
Constrained-guided mapping is proposed, a neuro-symbolic method with three stages: schema-grounded admissibility constraints with metadata mc =, where tau_c denotes the constraint type and delta_c provides executable relation and normalization logic, and constraint-restricted candidate generation with cascade relaxation to guarantee a nonempty feasible set under noise.
Findings support a hybrid paradigm in which AI augments, but does not replace, health economists in value assessment and formulary decision support within managed care settings.
Data Management
What if pathology foundation models could do more with less? GigaPath-Flash and GigaTIME-Flash cut computational demands while maintaining strong performance, opening the door to larger studies and broader exploration. The post GigaPath-Flash and GigaTIME-Flash: Toward population-scale discovery with efficient pathology foundation models appeared first on Microsoft Research.
With millions of users across the world, Julia has been used to conduct cutting-edge research and to design new drugs, jet engines, heat pumps, and more.
Earth AI
Health & Bioscience