TIBECG: Temporal Information Bottleneck for Lead-Specific Demographic Privacy in ECG Waveform Release
arXiv:2611.21694v1 Announce Type: new
Abstract: Transmitting an electrocardiogram (ECG) waveform can expose demographic information beyond an intended diagnostic task. We evaluate TIBECG, an architectural temporal information bottleneck, on individual leads extracted from hospital twelve-lead recor...
DeltaHarness: Self-Evolving Harnesses for Resource-Adaptive Tool Execution
arXiv:2611.21688v1 Announce Type: new
Abstract: Tool-using agents rely on execution harnesses to organize and schedule tool operations. The same task and resource description can conceal runtime conditions that reverse the latency ordering of valid execution plans. Resolving this uncertainty requir...
AdaFlowLite: Scalable and Non-blocking Inference on Asynchronous Mobile Data
arXiv:2611.21691v1 Announce Type: new
Abstract: The rise of mobile devices equipped with numerous sensors, such as LiDAR and cameras, has driven the adoption of multi-modal deep intelligence for distributed sensing tasks, such as smart cabins and driving assistance. However, the arrival time of mob...
Layerwise Learning-Rate Scaling in Vision Networks: Effective-Rate Identification, Power-Mean Aggregation, and Budget Dependence
arXiv:2611.21696v1 Announce Type: new
Abstract: Layerwise learning-rate scaling assigns different learning rates to different parts of a network. Although established in fine-tuning pretrained models, its effects in vision networks trained from scratch are commonly assessed by training each candida...
NEBench R1: a Reasoning Benchmark for Evaluating Nuclear Domain Competence in Large Language Models
arXiv:2611.21682v1 Announce Type: new
Abstract: Large language models (LLMs) are increasingly being explored for knowledge-intensive and safety-critical applications in nuclear engineering. However, existing nuclear-domain benchmarks provide limited content coverage or focus primarily on answer acc...
How Much of SAR Vehicle Recognition Is the Vehicle? Background Shortcuts, Angle Gaps and Partial Libraries on MSTAR and ATRNet-STAR
arXiv:2611.21827v1 Announce Type: new
Abstract: Deep networks score near 99% on the MSTAR vehicle benchmark, and the newer ATRNet-STAR benchmark was built to be harder. We ask how much of these scores comes from the vehicle itself. With a modern network (ConvNeXt-Small, 3 seeds) we first reproduce ...
Auditors Struggle to Catch Sabotage In End-to-End Transcripts of AI Safety Research
arXiv:2611.21832v1 Announce Type: new
Abstract: As AI systems are increasingly used to automate AI R&D, misaligned systems could sabotage research intended to improve model safety and alignment. To stress-test defenses against such sabotage, we use a human expert as a proxy saboteur on AI safety re...
arXiv:2611.21885v1 Announce Type: new
Abstract: Language models are trained once, frozen and deployed: improving one again takes another training run, and a prompt, a scaffold or extra computation at inference is gone once the task ends. A frontier model has been shown to edit the weights of a much...
Chain-of-Thought Controllability Can Reduce Chain-of-Thought Monitorability
arXiv:2611.21799v1 Announce Type: new
Abstract: Chain-of-thought (CoT) monitoring is a promising tool for detecting hidden goals, off-task behavior, and other failures in AI agents. One risk is that models could deliberately shape their reasoning to make it opaque to monitors. We stress-test this r...
What can AI enable in medicine, and how do we get there? Microsoft Research lead Doug Burger, Microsoft Principal Research Manager Ava Amini, Microsoft Chief Scientific Officer Eric Horvitz, and physician-scientist Eric Topol of Scripps Research explore these questions and more.
The post Can AI help...
Video-as-Experience: Learning Computer Use via Audio-Visual Understanding
arXiv:2611.18747v1 Announce Type: new
Abstract: Multimodal large language models (LLMs) have enabled capable computer-use agents that perceive graphical user interfaces, reason over instructions, and execute actions. However, existing approaches to learning from video demonstrations typically conve...
Agents Cannot Hand Over What They Have Not Yet Understood
arXiv:2611.18642v1 Announce Type: new
Abstract: Agents that work beyond a single context window, or in teams, hand their work to a successor as a short written summary. Evaluation has focused on whether an agent can solve a task, and relatively little is known about how much of what an agent has le...
TokenEconomizer: Learning from Experience to Make AI Coding Agents More Efficient
AI coding agents are increasingly help developers fix bugs and build features largely on their own, but every interaction with their models consumes tokens. In each session, an agent sends instructions, tool descriptions, and session history to its model. Most efforts to reduce this cost focus on in...
6 Washington companies proving AI agents can pass an audit
Washington pairs the federal government's agentic ambitions with enterprises shipping multi-agent systems at scale. Here is who builds what around the Beltway.
Tabular Population Priors as Structured Retrieval for EHR Medication Set Prediction
arXiv:2611.17341v1 Announce Type: new
Abstract: Medication prediction from electronic health records is a fixed vocabulary structured prediction problem, yet recent clinical RAG systems often treat it like open ended case based reasoning: retrieve similar patient admissions and ask an LLM to infer ...
When Should LLM Agents Forget? Memory Filtering Under Query-Critical Resolver Errors
arXiv:2611.17338v1 Announce Type: new
Abstract: LLM memory systems often delete facts classified as stale. We show that this hard-filtering step can reduce downstream accuracy when resolver errors delete query-critical facts. Across three models, a controlled sensitivity intervention moves hard-dro...
Does the Ranking Hold? A Practitioner-Facing Audit of Hallucination Detectors on Small, Quantized, On-Device LLMs
arXiv:2611.17339v1 Announce Type: new
Abstract: Hallucination detection for large language models is dominated by methods evaluated on 7B-70B research models with abundant sampling budgets and exact-match or ROUGE correctness labels. The small models practitioners actually ship on-device, Phi-3-min...
LoRAWeaver: Directional Adapter Routing in LoRA Representation Space
arXiv:2611.17340v1 Announce Type: new
Abstract: Large language models are expensive to update through full retraining. Parameter-efficient fine-tuning reduces the cost of adaptation, but repeated adaptation creates a growing pool of task-specific LoRA adapters. This pool complicates adapter selecti...
arXiv:2611.17495v1 Announce Type: new
Abstract: Chain-of-thought (CoT) is currently a valuable tool for overseeing AI models, but some architectural shifts could significantly reduce CoT monitorability. We operationalize a measure that serves as a proxy for the amount of unverbalized serial cogniti...
Unfinished-Bench: Measuring Rule Violations by Persistent Agents in Realistic Work Environments
arXiv:2611.17460v1 Announce Type: new
Abstract: AI agents are increasingly equipped with persistence toward task completion to run delegated work autonomously for long periods, under many unstated constraints from internal documents and other people's authority. When completion requires violating a...
Diffusion Tree Sampling for Inference Time Adaptation of Crystal Generative Models
arXiv:2611.17462v1 Announce Type: new
Abstract: Diffusion-based generative models have become strong priors over plausible crystal structures, but turning that prior into promising candidates requires steering toward several conflicting properties at once. Fine-tuning is expensive and produces a ne...
WETW: Learning Pair-Specific Weighted Euclidean Distances for Efficient DTW Search
arXiv:2611.17412v1 Announce Type: new
Abstract: Dynamic Time Warping (DTW) is the standard distance for similarity search over time series that are misaligned in time, but its cost is quadratic in the series length and lower bounds often prune few candidates, so search over large collections of lon...
Latent 3D representations have significantly expanded the capabilities of modern shape modeling, enabling compact encoding of 3D scenes and supporting powerful 3D generative models. Despite this progress, current representations struggle to capture the fine details of complex objects, fundamentally ...