From prediction to decision: Smaller models will reshape how we build AI
The bottleneck in AI systems isn't model intelligence anymore, but decision agility. Mallika Rao, engineering manager and former Netflix infrastructure lead, breaks down why smaller, task-specific models are outrunning foundational ones, and what that means for every team building at scale.
Closing the Social-Semantic Gap: SPSD for Edge-Based Prompt Compression in Cloud LLM Inference
arXiv:2606.19364v1 Announce Type: new
Abstract: The prefill stage of Large Language Model (LLM) inference is a growing contributor to cloud-scale energy cost. Many consumer-support and conversational prompts contain social scaffolding: politeness markers, apologetic preamble, repetition, and rappor...
Performance Analysis and Optimization of 3D Generative Diffusion Models across GPU Architectures
arXiv:2606.19365v1 Announce Type: new
Abstract: Diffusion models have become essential for high-fidelity 3D MRI synthesis, yet their deployment remains constrained by substantial GPU resource demands arising from hundreds of U-Net evaluations per sample and a highly heterogeneous kernel behavior. T...
Information Lattice Learning as Probabilistic Graphical Model Structure Learning
arXiv:2606.19366v1 Announce Type: new
Abstract: Information lattice learning (ILL) learns interpretable rules of a signal by alternately projecting the signal onto a partition lattice that encodes a hierarchy of abstractions and lifting selected rules back to the signal domain. When the signal is a...
Deontic Policies for Runtime Governance of Agentic AI Systems
arXiv:2606.19464v1 Announce Type: new
Abstract: Autonomous agentic AI systems driven by Large Language Models (LLMs) introduce a new class of security, privacy, and compliance challenges: an agent that can invoke tools, manipulate data, install software, and coordinate with peer agents across organ...
Measuring Curriculum Alignment across Topical Coverage, Competency, and Cognitive Depth: A Longitudinal Framework Applied to CS2013 and CS2023
arXiv:2606.19469v1 Announce Type: new
Abstract: Undergraduate computer science is governed by international curricular guidelines revised about once a decade, yet programs lack a reliable, reproducible way to measure how completely they cover the current guidelines and how that coverage shifts when...
Diffusion Language Models: An Experimental Analysis
arXiv:2606.19475v1 Announce Type: new
Abstract: Large Language Models (LLMs) have revolutionized language modeling through autoregressive generation, enabling strong performance across a wide range of tasks. Recently, Diffusion Language Models (DLMs) have emerged as an alternative paradigm that gen...
arXiv:2606.19494v1 Announce Type: new
Abstract: Multi-agent LLM deliberation, where agents exchange and revise answers over several rounds, is increasingly used to improve reasoning and accuracy, yet how and why it works is rarely modelled. Such deliberation mirrors how humans reach decisions. As s...
DeXposure-Claw: An Agentic System for DeFi Risk Supervision
arXiv:2606.19501v1 Announce Type: new
Abstract: Decentralized finance exposes supervisors to fast-moving, networked credit risks. General-purpose LLM agents fit this setting poorly: they over-read weak evidence and recommend high-stakes interventions, while existing evaluations offer no regulator-a...
SpaceX wants to build AI data centers in space. Will it work?
The race to build data centers in space is gaining momentum as AI drives unprecedented demand for computing power. Orbital facilities could tap into abundant solar energy and avoid many of the environmental challenges faced on Earth. Yet space remains a harsh and expensive place to operate, with maj...
Gaussian Mixture Attention: Linear-Time Sequence Mixing via Probabilistic Latent Routing
arXiv:2606.18283v1 Announce Type: new
Abstract: The dense token-to-token interaction pattern of standard dot-product attention remains a central bottleneck in scaling Transformer architectures to long contexts. We introduce \textbf{Gaussian Mixture Attention (GMA)}, a probabilistic attention-style ...
Breaking the Solver Bottleneck: Training Task Generators at the Learnable Frontier
arXiv:2606.18284v1 Announce Type: new
Abstract: The limiting resource for training agents via reinforcement learning (RL) is increasingly frontier task supply: valid, solvable tasks just difficult enough to train the current model. As reasoning and agentic models improve, fixed task distributions s...
CODEBLOCK: Learning to Supervise Code at the Right Granularity
arXiv:2606.18286v1 Announce Type: new
Abstract: Supervised fine-tuning of code LLMs typically applies uniform cross-entropy loss to all response tokens, implicitly assuming that every token provides equally useful learning signal. Recent token-level selection methods challenge this assumption in na...
A Link between Shock-wave Theory and Symmetry-reduced Stochastic Gradient Descent for Artificial Neural Networks
arXiv:2606.18303v1 Announce Type: new
Abstract: We develop a mathematically explicit link between shock-wave theory and the symmetry-quotiented learning dynamics of stochastic gradient descent, drawing on differential geometry, Lie group theory, and fluid mechanics. Specifically, after quotienting ...
NAVI-Orbital: First In-Orbit Demonstration of a Zero-Shot Vision-Language Model for Autonomous Earth Observation
arXiv:2606.18271v1 Announce Type: new
Abstract: As Earth Observation data generation outpaces downlink bandwidth and human-in-the-loop processing, a widening gap has emerged between onboard collection and actionable ground intelligence. This paper presents NAVI-Orbital, a software system deployed o...
CaVe-VLM-CoT: An Interpretable Vision-Language Model Framework
arXiv:2606.18385v1 Announce Type: new
Abstract: Vision-Language Models (VLMs) remain prone to hallucinations, producing fluent but visually unfaithful outputs. Existing chain-of-thought and retrieval-augmented methods only partially address this, as they neither enforce step-level citation groundin...
Searching for Synergy in Shared Workspace Human-AI Collaboration
arXiv:2606.18413v2 Announce Type: new
Abstract: Automated AI agents are increasingly capable, yet many scientific and professional tasks require human judgment and contextual expertise. We study shared-workspace human-AI teams, where AI agents and human collaborators must coordinate responsibilitie...
arXiv:2606.18543v1 Announce Type: new
Abstract: Language model agents are becoming proficient executors at isolated, short-horizon tasks such as software engineering and customer service. Yet real-world challenges require a combination of sophisticated skills that remain largely untested in agents:...
DeFAb: A Verifiable Benchmark for Defeasible Abduction in Foundation Models
arXiv:2606.18557v1 Announce Type: new
Abstract: A rule-based logic solver resolves every instance in our benchmark in under 50 microseconds with 100% accuracy; the best frontier language model reaches 65% at best and drops to 23.5% under rendering-robust evaluation (worst case over four surface ren...
Is your most capable AI agent also your biggest data leak?
A Microsoft and Huazhong University benchmark tested GPT-4o, GPT-5, Grok-3, and others on realistic enterprise data scenarios. Privacy violation rates hit 50.9%. More capable models made it worse, and the fix has nothing to do with model selection...
Correct When Paired, Wrong When Split: Decoupling and Editing Modality-Specific Neurons in MLLMs
arXiv:2606.17057v1 Announce Type: new
Abstract: Although Knowledge Editing provides an efficient mechanism for updating the knowledge of Multimodal Large Language Models (MLLMs), we find that current paradigms still suffer from an important yet remain underexplored issue : editing decoupling failur...
Informative Missingness to Generate Irregular Clinical Time Series
arXiv:2606.17106v1 Announce Type: new
Abstract: Laboratory tests in electronic health records are collected irregularly, and the absence of a test order can be as informative as the measurement itself. Such missingness reflects clinicians' decisions and patient physiology, making it important to mo...
Models Take Notes at Prefill: KV Cache Can Be Editable and Composable
arXiv:2606.17107v1 Announce Type: new
Abstract: Prefix caching reuses prefill only across an exactly shared prefix, so one changed field invalidates the entire downstream cache. Yet overwriting the field's own key/value vectors and reusing the rest leaves the model acting on the old value. The reas...
The Critical Role of Model Selection in Causal Inference: A Comparative Analysis of Classification Models within the InferBERT Framework for Pharmacovigilance
arXiv:2606.17113v1 Announce Type: new
Abstract: Distinguishing causal adverse drug events (ADEs) from spurious correlations remains a central challenge in pharmacovigilance. The InferBERT framework integrates transformer models with Do-calculus, but its success hinges on the underlying classificati...