Meta-TTRL: A Metacognitive Framework for Self-Improving Test-Time Reinforcement Learning in Unified Multimodal Models
arXiv:2603.15724v1 Announce Type: new Abstract: Existing test-time scaling (TTS) methods for unified multimodal models (UMMs) in text-to-image (T2I) generation primarily rely on search or sampling strategies that produce only instance-level improvements, limiting the ability to learn from prior inferences and...
Longitudinal Risk Prediction in Mammography with Privileged History Distillation
arXiv:2603.15814v1 Announce Type: new Abstract: Breast cancer remains a leading cause of cancer-related mortality worldwide. Longitudinal mammography risk prediction models improve multi-year breast cancer risk prediction based on prior screening exams. However, in real-world clinical practice, longitudinal histories are often...
Hypothesis Class Determines Explanation: Why Accurate Models Disagree on Feature Attribution
arXiv:2603.15821v1 Announce Type: new Abstract: The assumption that prediction-equivalent models produce equivalent explanations underlies many practices in explainable AI, including model selection, auditing, and regulatory evaluation. In this work, we show that this assumption does not hold. Through a large-scale...
Counteractive RL: Rethinking Core Principles for Efficient and Scalable Deep Reinforcement Learning
arXiv:2603.15871v1 Announce Type: new Abstract: Following the pivotal success of learning strategies to win at tasks, solely by interacting with an environment without any supervision, agents have gained the ability to make sequential decisions in complex MDPs. Yet, reinforcement learning...
Game-Theory-Assisted Reinforcement Learning for Border Defense: Early Termination based on Analytical Solutions
arXiv:2603.15907v1 Announce Type: new Abstract: Game theory provides the gold standard for analyzing adversarial engagements, offering strong optimality guarantees. However, these guarantees often become brittle when assumptions such as perfect information are violated. Reinforcement learning (RL), by contrast, is adaptive...
Discovery of interaction and diffusion kernels in particle-to-mean-field multi-agent systems
arXiv:2603.15927v1 Announce Type: new Abstract: We propose a data-driven framework to learn interaction kernels in stochastic multi-agent systems. Our approach aims at identifying the functional form of nonlocal interaction and diffusion terms directly from trajectory data, without any a priori...
Deriving Hyperparameter Scaling Laws via Modern Optimization Theory
arXiv:2603.15958v1 Announce Type: new Abstract: Hyperparameter transfer has become an important component of modern large-scale training recipes. Existing methods, such as muP, primarily focus on transfer between model sizes, with transfer across batch sizes and training horizons often relying on...
W2T: LoRA Weights Already Know What They Can Do
arXiv:2603.15990v1 Announce Type: new Abstract: Each LoRA checkpoint compactly stores task-specific updates in low-rank weight matrices, offering an efficient way to adapt large language models to new tasks and domains. In principle, these weights already encode what the adapter does...
The Importance of Being Smoothly Calibrated
arXiv:2603.16015v1 Announce Type: new Abstract: Recent work has highlighted the centrality of smooth calibration [Kakade and Foster, 2008] as a robust measure of calibration error. We generalize, unify, and extend previous results on smooth calibration, both as a robust calibration...
Residual Stream Duality in Modern Transformer Architectures
arXiv:2603.16039v1 Announce Type: new Abstract: Recent work has made clear that the residual pathway is not mere optimization plumbing; it is part of the model's representational machinery. We agree, but argue that the cleanest way to organize this design space...
MDM-Prime-v2: Binary Encoding and Index Shuffling Enable Compute-optimal Scaling of Diffusion Language Models
arXiv:2603.16077v1 Announce Type: new Abstract: Masked diffusion models (MDM) exhibit superior generalization when learned using a Partial masking scheme (Prime). This approach converts tokens into sub-tokens and models the diffusion process at the sub-token level. We identify two limitations of...
The remaining questions after the Supreme Court’s tariffs ruling
Last month, the Supreme Court ruled that the International Emergency Economic Powers Act, a 1977 law giving the president the power to regulate commerce during national emergencies created by foreign […]The postThe remaining questions after the Supreme Court’s tariffs rulingappeared...
Apple can delist apps "with or without cause," judge says in loss for Musi app
Judge tosses Musi case against Apple, sanctions lawyers for "mak[ing] up facts."
Arizona indicts prediction market Kalshi for running illegal gambling operation
Desert state becomes first to file criminal case against prediction platform.
The Pentagon is developing alternatives to Anthropic, report says
After their dramatic falling-out, it doesn't seem as though Anthropic and the Pentagon are getting back together.
BuzzFeed debuts AI slop apps in bid for new revenue
BuzzFeed unveiled new AI-powered social apps at SXSW, but its demos drew muted reactions.
Google’s Personal Intelligence feature is expanding to all US users
Personal Intelligence allows Google's AI assistant to tap into your Google ecosystem, such as Gmail and Google Photos, to provide more tailored responses.
OpenAI expands government footprint with AWS deal, report says
OpenAI has reportedly signed a partnership with AWS to sell its AI systems to the U.S. government for classified and unclassified work, marking an expansion beyond its Pentagon deal last month.
AI’s ‘boys’ club’ could widen the wealth gap for women, says Rana el Kaliouby
AI investor Rana el Kaliouby warns that if women are shut out of AI funding and leadership, the consequences will be grim.
World launches tool to verify humans behind AI shopping agents
As AI agents take the reins for online shoppers, Sam Altman's unconventional startup is looking to expand its verification offerings to support agentic commerce.
Niv-AI exits stealth to wring more power performance out of GPUs
The company raised $12 million in seed funding to measure and manage GPU power surges.
Gamma adds AI image-generation tools in bid to take on Canva and Adobe
The company's new product, called Gamma Imagine, will let users employ text prompts to create brand-specific assets like interactive charts and visualizations, marketing collateral, social graphics, and infographics.
Picsart now allows creators to ‘hire’ AI assistants through agent marketplace
Picsart's AI agent marketplace will launch with four agents, then add more agents each week.
Her Fundamental Right To Procreate: The Unconstitutionality Of Abortion Bans
As she was wheeled into surgery, Amber Thurman said to her mother, “Promise me you’ll take care of my son.” She was suffering a rare complication from the abortion pill that she was legally prescribed at nine weeks of pregnancy....
Why Grokking Takes So Long: A First-Principles Theory of Representational Phase Transitions
arXiv:2603.13331v1 Announce Type: new Abstract: Grokking is the sudden generalization that appears long after a model has perfectly memorized its training data. Although this phenomenon has been widely observed, there is still no quantitative theory explaining the length of the...
Traffic and weather driven hybrid digital twin for bridge monitoring
arXiv:2603.14028v1 Announce Type: new Abstract: A hybrid digital twin framework is presented for bridge condition monitoring using existing traffic cameras and weather APIs, reducing reliance on dedicated sensor installations. The approach is demonstrated on the Peace Bridge (99 years in...
Formal Abductive Explanations for Navigating Mental Health Help-Seeking and Diversity in Tech Workplaces
arXiv:2603.14007v1 Announce Type: new Abstract: This work proposes a formal abductive explanation framework designed to systematically uncover rationales underlying AI predictions of mental health help-seeking within tech workplace settings. By computing rigorous justifications for model outputs, this approach enables principled...
vla-eval: A Unified Evaluation Harness for Vision-Language-Action Models
arXiv:2603.13966v1 Announce Type: new Abstract: Vision Language Action VLA models are typically evaluated using per benchmark scripts maintained independently by each model repository, leading to duplicated code, dependency conflicts, and underspecified protocols. We present vla eval, an open source evaluation...
The ARC of Progress towards AGI: A Living Survey of Abstraction and Reasoning
arXiv:2603.13372v1 Announce Type: new Abstract: The Abstraction and Reasoning Corpus (ARC-AGI) has become a key benchmark for fluid intelligence in AI. This survey presents the first cross-generation analysis of 82 approaches across three benchmark versions and the ARC Prize 2024-2025...
Prompt Complexity Dilutes Structured Reasoning: A Follow-Up Study on the Car Wash Problem
arXiv:2603.13351v1 Announce Type: new Abstract: In a previous study [Jo, 2026], STAR reasoning (Situation, Task, Action, Result) raised car wash problem accuracy from 0% to 85% on Claude Sonnet 4.5, and to 100% with additional prompt layers. This follow-up asks:...