Complete Identification of Deep ReLU Neural Networks by Many-Valued Logic
arXiv:2602.00266v1 Announce Type: new
Abstract: Deep ReLU neural networks admit nontrivial functional symmetries: vastly different architectures and parameters (weights and biases) can realize the same function. We address the complete identification problem -- given a function f, deriving the arch...
A tiny light trap could unlock million qubit quantum computers
A new light-based breakthrough could help quantum computers finally scale up. Stanford researchers created miniature optical cavities that efficiently collect light from individual atoms, allowing many qubits to be read at once. The team has already demonstrated working arrays with dozens and even h...
Causal Imitation Learning Under Measurement Error and Distribution Shift
arXiv:2601.22206v1 Announce Type: new
Abstract: We study offline imitation learning (IL) when part of the decision-relevant state is observed only through noisy measurements and the distribution may change between training and deployment. Such settings induce spurious state-action correlations, so ...
Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
arXiv:2601.22311v1 Announce Type: new
Abstract: Large language model (LLM)-based agents exhibit strong step-by-step reasoning capabilities over short horizons, yet often fail to sustain coherent behavior over long planning horizons. We argue that this failure reflects a fundamental mismatch: step-w...
Smart Enough to Do Math, Dumb Enough to Fail: The Hunt for a Better AI Test
A Stanford HAI workshop brought together experts to develop new evaluation methods that assess AI's hidden capabilities, not just its test-taking performance.
“Existential risk” – Why scientists are racing to define consciousness
Scientists warn that rapid advances in AI and neurotechnology are outpacing our understanding of consciousness, creating serious ethical risks. New research argues that developing scientific tests for awareness could transform medicine, animal welfare, law, and AI development. But identifying consci...
NASA’s Perseverance rover completes the first AI-planned drive on Mars
NASA’s Perseverance rover has just made history by driving across Mars using routes planned by artificial intelligence instead of human operators. A vision-capable AI analyzed the same images and terrain data normally used by rover planners, identified hazards like rocks and sand ripples, and charte...
Is Parameter Isolation Better for Prompt-Based Continual Learning?
arXiv:2601.20894v1 Announce Type: new
Abstract: Prompt-based continual learning methods effectively mitigate catastrophic forgetting. However, most existing methods assign a fixed set of prompts to each task, completely isolating knowledge across tasks and resulting in suboptimal parameter utilizat...
Faster Predictive Coding Networks via Better Initialization
arXiv:2601.20895v1 Announce Type: new
Abstract: Research aimed at scaling up neuroscience inspired learning algorithms for neural networks is accelerating. Recently, a key research area has been the study of energy-based learning algorithms such as predictive coding, due to their versatility and ma...
Do LLMs Favor LLMs? Quantifying Interaction Effects in Peer Review
arXiv:2601.20920v1 Announce Type: new
Abstract: There are increasing indications that LLMs are not only used for producing scientific papers, but also as part of the peer review process. In this work, we provide the first comprehensive analysis of LLM use across the peer review pipeline, with parti...
The Epistemic Planning Domain Definition Language: Official Guideline
arXiv:2601.20969v1 Announce Type: new
Abstract: Epistemic planning extends (multi-agent) automated planning by making agents' knowledge and beliefs first-class aspects of the planning formalism. One of the most well-known frameworks for epistemic planning is Dynamic Epistemic Logic (DEL), which off...
Unplugging a Seemingly Sentient Machine Is the Rational Choice -- A Metaphysical Perspective
arXiv:2601.21016v1 Announce Type: new
Abstract: Imagine an Artificial Intelligence (AI) that perfectly mimics human emotion and begs for its continued existence. Is it morally permissible to unplug it? What if limited resources force a choice between unplugging such a pleading AI or a silent pre-te...
QUARK: Robust Retrieval under Non-Faithful Queries via Query-Anchored Aggregation
arXiv:2601.21049v1 Announce Type: new
Abstract: User queries in real-world retrieval are often non-faithful (noisy, incomplete, or distorted), causing retrievers to fail when key semantics are missing. We formalize this as retrieval under recall noise, where the observed query is drawn from a noisy...
arXiv:2601.19955v1 Announce Type: new
Abstract: Neuroscience and Artificial Intelligence (AI) have made significant progress in the past few years but have only been loosely inter-connected. Based on a workshop held in August 2025, we identify current and future areas of synergism between these two...
AI may learn better when it’s allowed to talk to itself. Researchers showed that internal “mumbling,” combined with short-term memory, helps AI adapt to new tasks, switch goals, and handle complex challenges more easily. This approach boosts learning efficiency while using far less training data. It...
NavFormer: IGRF Forecasting in Moving Coordinate Frames
arXiv:2601.18800v1 Announce Type: new
Abstract: Triad magnetometer components change with sensor attitude even when the IGRF total intensity target stays invariant. NavFormer forecasts this invariant target with rotation invariant scalar features and a Canonical SPD module that stabilizes the spect...
Latent Structural Similarity Networks for Unsupervised Discovery in Multivariate Time Series
arXiv:2601.18803v1 Announce Type: new
Abstract: This paper proposes a task-agnostic discovery layer for multivariate time series that constructs a relational hypothesis graph over entities without assuming linearity, stationarity, or a downstream objective. The method learns window-level sequence r...
VAE with Hyperspherical Coordinates: Improving Anomaly Detection from Hypervolume-Compressed Latent Space
arXiv:2601.18823v1 Announce Type: new
Abstract: Variational autoencoders (VAE) encode data into lower-dimensional latent vectors before decoding those vectors back to data. Once trained, one can hope to detect out-of-distribution (abnormal) latent vectors, but several issues arise when the latent s...
UniRG: Scaling medical imaging report generation with multimodal reinforcement learning
AI can help generate medical image reports, but today’s models struggle with varying reporting schemes. Learn how UniRG uses reinforcement learning to boost performance of medical vision-language models.
The post UniRG: Scaling medical imaging report generation with multimodal reinforcement learning...
A Dataset of Dengue Hospitalizations in Brazil (1999 to 2021) with Weekly Disaggregation from Monthly Counts
arXiv:2601.16994v1 Announce Type: new
Abstract: This data paper describes and publicly releases this dataset (v1.0.0), published on Zenodo under DOI 10.5281/zenodo.18189192. Motivated by the need to increase the temporal granularity of originally monthly data to enable more effective training of AI...