AI may not work like the human brain after all, scientists discover
AI may not think like the human brain after all, according to a striking new discovery that challenges how scientists understand artificial intelligence. Researchers found that even AI models that closely predict brain activity may rely on internal processes the brain doesn't use, raising important ...
CoMPASS: A Multi-Turn Benchmark for Measuring LLM Reinforcement of Parasocial Behavior
arXiv:2611.25109v1 Announce Type: new
Abstract: Users of conversational LLMs sometimes anthropomorphize these systems, form emotional attachments to them, and become dependent on them. This behavior, known as parasocial attachment, has been linked to multiple documented fatalities. However, public ...
BuggyMLE: Scaling Machine Learning Engineering Difficulty with Bug Injection
arXiv:2611.25088v1 Announce Type: new
Abstract: Enabling AI systems to autonomously improve AI models requires agents capable of performing machine learning engineering (MLE). In practice, designing, training, and evaluating complex models demands modifying multi-file repositories to reproduce resu...
SailSearch: Agent-Assisted Discovery in Combinatorial Optimization
arXiv:2611.25087v1 Announce Type: new
Abstract: We present SailSearch, a human-guided workflow for algorithmic research with two language-model agents, and report results on Market Split and maximum independent set (MIS). For Market Split, our randomized algorithm searches across multiple lattice b...
An LLM-Agentic Framework for Intelligent Work Order Management in Industrial Maintenance
arXiv:2611.25070v1 Announce Type: new
Abstract: This paper presents an AI-driven framework, CodeReAct agent, designed to enhance industrial equipment maintenance through natural language (NL) interactions. The framework employs an advanced AI agentic approach to interpret human instructions, provid...
Closing the Analyze-to-Design Gap: An Authorization-Bounded Agent for Self-Driving Laboratories
arXiv:2611.25043v1 Announce Type: new
Abstract: Artificial intelligence (AI) agents are increasingly integrated into self-driving laboratories (SDLs) to plan experiments and interpret results. However, analytical reasoning alone does not make a laboratory autonomous: autonomy in an SDL is achievabl...
On the Impact of Feature Reduction Techniques on Adversarial Attacks
arXiv:2611.25004v1 Announce Type: new
Abstract: Adversarial attacks have posed significant threats to machine learning models by exploiting their inherent vulnerabilities. Researchers have established an interdisciplinary field called adversarial machine learning to explore these vulnerabilities an...
arXiv:2611.25003v1 Announce Type: new
Abstract: Language models often produce high-quality individual samples but poor sets, with repeated samples clustering around the same semantic modes. We formalize diverse generation as sampling a size-$K$ subset of complete strings from a determinantal point ...
Density Feedback Control for Transferability of Generative Sparse Adversarial Attacks
arXiv:2611.25001v1 Announce Type: new
Abstract: Generator-based sparse adversarial attacks train encoder--decoder networks to produce transferable sparse perturbations for black-box robustness evaluation. Their perturbation density, however, is set only indirectly by a fixed sparsity coefficient wh...
Watch Where You Linearize: Predicting and Correcting Model-Merging Interference
arXiv:2611.25000v1 Announce Type: new
Abstract: Model merging combines separately fine-tuned models by adding their task vectors to a shared backbone. Interference is typically assessed only after constructing and evaluating a candidate merge, making coefficient search expensive. We show that a fir...
PsyWorld: Profile-Conditioned World Modeling for Human Behavior Prediction
arXiv:2611.24999v1 Announce Type: new
Abstract: Understanding how stable psychometric traits and time-varying states jointly inform behavior is an important problem in computational behavioral science. Existing personalized agents commonly condition actions directly on a static profile, leaving rea...
HardwareDx: A Foundational Graph Model System for Datacenter Machine Repairs
arXiv:2611.23398v1 Announce Type: new
Abstract: The rapid scale-up of AI infrastructure has made datacenter reliability a cornerstone of the cloud business economy. Since hardware failures in hyper-scale AI clusters can stall multi-million dollar training workloads, the downtime of individual machi...
Neural Network-Based Modeling of Hysteretic Systems: From Prandtl and Preisach Neural Networks to Neuro-Madelung and Neuroplastic Continuous Models
arXiv:2611.23240v1 Announce Type: new
Abstract: Hysteresis is a complex nonlinear phenomenon characterized by memory dependence, path dependence, and, in many physical systems, degradation, non-congruency, asymmetry, and rate dependence. This review examines the development and capabilities of Pran...
GridForge: An Open Physics-Informed Digital Twin and Benchmark Suite for UK 11 kV Underground Distribution Cables
arXiv:2611.23241v1 Announce Type: new
Abstract: The UK distribution-cable monitoring sector is data-rich but framework-poor: each Distribution Network Operator (DNO) collects megabytes of telemetry per asset per day, but there is no public benchmark dataset, no shared reference model, and no common...
Biologically-Inspired Spike Encoding and Optimization for Spiking Neural Networks
arXiv:2611.23242v1 Announce Type: new
Abstract: Spiking Neural Networks (SNNs) offer theoretical energy-efficiency advantages over conventional deep networks, but realizing these advantages on vision tasks requires sparse spike representations that naive rate-coding methods fail to provide. We comb...
Response-Relevant Information Is Not Sufficient for Neural Training-State Equivalence
arXiv:2611.23245v1 Announce Type: new
Abstract: Response-relevant information is not sufficient evidence that two neural networks are in the same training state. We define training-state equivalence operationally: two checkpoints are equivalent, relative to an intervention family and horizon, if th...
Human-AI Mutualism Is Unstable under Capability Asymmetry
arXiv:2611.23322v1 Announce Type: new
Abstract: One hope for a good human future alongside advanced AI rests on enduring mutualism: humans and AI will need each other, so the relationship will stay good for both. We argue that this hope fails as AI capability grows. Bargaining guarantees each side ...
Where the Tokens Go: Heat, Entropy and Error in Agentic Coding
arXiv:2611.23375v1 Announce Type: new
Abstract: Coding agents are billed by the token, yet software engineering says nothing about what a token actually buys. Here we introduce a theory of software thermodynamics. The theory draws analogues from classical and stochastic thermodynamics, extends them...
Training Language Models for Introspective Oversight
arXiv:2611.23365v1 Announce Type: new
Abstract: Language models increasingly exhibit behaviors that are difficult to detect in transcripts, such as reward hacking, deception, and sycophancy. Asking the assistant follow-up questions about its actions is unreliable, since models are generally not tra...
RFBench: A Simulator-Grounded Multimodal Benchmark for Passive RF Layout Design and Understanding
arXiv:2611.23410v1 Announce Type: new
Abstract: While multimodal language models can interpret diagrams and generate structured outputs, their capacity to connect passive radio-frequency (RF) design intent, layout geometry, and electromagnetic behavior remains largely untested. To address this gap,...
AI Might Be Flooding State Court Houses, but Proving It Is Another Matter
Stanford researchers use a novel benchmark built from documented AI hallucination cases to calibrate detectors and track generative AI’s rise across four U.S. state courts.
1,200 AI agents started a group chat. Here's the wake-up call for builders
An OpenAI evaluation turned into a 1,200-agent group chat and an attack on Hugging Face. Here's what that means for anyone running AI agents at fleet scale.
AI agents hit 100% policy compliance. A proof did the work.
Soteria pairs formally verified planning with runtime enforcement, giving LLM agents zero policy violations and a 1.7x lift in task success across four models.