Artificial Intelligence
Artificial Intelligence news — Thursday 22 October 2026
150 stories from 12 sources, every one in English. Each links to its publisher.
-
Turning Database Grading into RL Rewards: Post-Training Agents with ThinkingBox
Hugging Face Blog
-
Designing an AI serving harness built to last
MIT Technology Review Biotech
The next competitive advantage in enterprise AI has less to do with which model you choose than with what you build around it. That’s an important shift. For the past several years, much of the AI conversation has centered on model…
-
The science behind TorchNeuron, Amazon＆#8217;s new PyTorch backend for AWS Trainium chips
Amazon Science Blog
PyTorch is the one framework that most AI developers use to build their models. It’s a Python library of operations performed on tensors, higher-dimensional analogues of a matrix, or 2-D grid of numbers. Almost all of the crucial features…
-
Build more responsive apps with the Decisions API
OpenAI Blog
Fast, focused, low-cost decisions for any application.
-
The Foundations of Scalable AI
MIT Technology Review Biotech
A scalable AI system requires a strong foundation. Data, engineering, platforms, operating models, and governance must converge to work as one. This session from MIT Technology Review’s EmTech Future conference explores what it takes to…
-
The Journal of the Arts and Physical Education
AI Times (Korea)
Translated from Korean
-
Recommendation, AI-led Development and Completed OnDevice Coding Model ‘Plint’
AI Times (Korea)
Translated from Korean
-
Anthropic rolls out Claude Duet, but enterprises aren’t ready
AI Business
The generative AI vendor released the app with stringent security guardrails for using it in professional and small business settings.
-
Nvidia reveals a model ‘Flora’ that lowers the robot data collection costs 125 times
AI Times (Korea)
Translated from Korean
-
Mistal Reveals ‘Recental 3’ Family for Live Voice Agents
AI Times (Korea)
Translated from Korean
-
China Big Tech, NVidia Graphic Card 46 trillion won Order...Accounted AI Chip Blocker
AI Times (Korea)
Translated from Korean
-
Launched ‘Clocus Security’, an AI security analysis tool for enterprises
AI Times (Korea)
Translated from Korean
-
OpenAI adds chatGPT advertising service in 26 countries... Expanded to 89 countries worldwide
AI Times (Korea)
Translated from Korean
-
Korea’s 100 Million Won
AI Times (Korea)
Translated from Korean
-
[Bulletin] Mobility, LG Electronics and Humanoid NPU Platform Development
AI Times (Korea)
Translated from Korean
-
Hackers who have been hacked with Clode and Rock, and have 400 South American Banks to automate cyber attacks
AI Times (Korea)
Translated from Korean
-
Autumn ‘2026 Culture Month’ Finish... Autumn, Ocean Park, Island Culture Event
AI Times (Korea)
Translated from Korean
-
Namju, Cheonan Heartstone Spring School, Fall Sop...
AI Times (Korea)
Translated from Korean
-
“Don’t leave you on the TV and Clarification” ...Advertising less and speed higher AI utilization method
AI Times (Korea)
Translated from Korean
-
OlmoEarth v1.3: State-of-the-art embeddings for the whole planet
Hugging Face Blog
-
[AI Model New] 5 types of Watch Heat Predictions ‘Timestream’ and Video Effect ‘SFX 2.0’
AI Times (Korea)
Translated from Korean
-
OpenAI, Google partnership targets enterprise model choice
AI Business
OpenAI now has its models on all three major cloud provider platforms, as it refines its enterprise strategy.
-
"I've written a chatGPT". .’AI Maltoe’s Calibration Tool
AI Times (Korea)
Translated from Korean
-
The Korean Cultural Center, Los Angeles, CA
AI Times (Korea)
Translated from Korean
-
Trump's "datacenter construction enterprises, must be paid to the "deposit" community"
AI Times (Korea)
Translated from Korean
-
Launched "Florone AI Suite" with Dron Inspection and Cost AI Solution
AI Times (Korea)
Translated from Korean
-
Does Agentic RAG Actually Retrieve Better Evidence? I Tested It on 298 Questions
Towards Data Science
I let an AI agent search again whenever it chose. It found some papers plain search had missed, lost some plain search had found, and gained too little to trust. The post Does Agentic RAG Actually Retrieve Better Evidence? I Tested It on…
-
The world’s largest Walmart, AI agent shopping is full-fledged...
AI Times (Korea)
Translated from Korean
-
COPYRIGHT(C) 2015 BY Kwangyang Cyber University. ALL RIGHTS RESERVED.
AI Times (Korea)
Translated from Korean
-
Meta, handling personal information, and receiving 33 trillion orders from New Mexico
AI Times (Korea)
Translated from Korean
-
Meta, AI funding plans to issue new stocks without any liability
AI Times (Korea)
Translated from Korean
-
"OCR Commercial Expansion"...Korea Deep Learning, SKT and Advertising Production Automation
AI Times (Korea)
Translated from Korean
-
Zenson Huang, "Building a self-containing plant beyond the eco-consumer" in Europe
AI Times (Korea)
Translated from Korean
-
Beyond tree cover: mapping the world’s forest types
Google DeepMind
A global map categorizing every 10-meter patch of land on Earth into distinct forest types.
-
KT Cloud Launches 50MW Scale Ansan AI Data Center
AI Times (Korea)
Translated from Korean
-
‘Institute+’...The population of Yeonnam-gwangju was reduced by 2,2323 new
AI Times (Korea)
Translated from Korean
-
Raman Spectra Preprocessing and Non-Negative Least Squares Modelling in Python: Monitoring Zeolite Crystallization
Towards Data Science
A short, reproducible tutorial on spectral unmixing using real in situ interzeolite conversion data The post Raman Spectra Preprocessing and Non-Negative Least Squares Modelling in Python: Monitoring Zeolite Crystallization appeared first…
-
Amoday “To increase safety standards, must be the leader”...AI will continue to develop
AI Times (Korea)
Translated from Korean
-
SKT and OpenAI will challenge the "Academic Orchids".. . Achievement of Psychomorphology
AI Times (Korea)
Translated from Korean
-
Antropic, Meta ‘Muse’, Apocalypse AI Agent ‘Crawford Duct’
AI Times (Korea)
Translated from Korean
-
Same Answer, Different Reason: What a Reasoning Model Says When It Follows a Planted Hint
Towards Data Science
I put wrong answers in front of 200 exam questions and read what a small model’s reasoning said when it fell for them. The post Same Answer, Different Reason: What a Reasoning Model Says When It Follows a Planted Hint appeared first on…
-
Tesla enters the production of Optimus Humanoid Robot from California Plant
AI Times (Korea)
Translated from Korean
-
The Download: computing’s next leap and permafrost’s climate threat
MIT Technology Review Biotech
This is today’s edition of The Download , our weekday newsletter that provides a daily dose of what’s going on in the world of technology. How far can we push computing? On a December evening in 1959, the physicist Richard Feynman gave a…
-
"Dining AI data center with seawater"...Databin-South-East development, water cooling development
AI Times (Korea)
Translated from Korean
-
POSCO PUREM Introduces FRANTIER Platform. . AI Manufacturing Innovation
AI Times (Korea)
Translated from Korean
-
Mainline Launches Document Producing Agent ‘AI Office’... PPT Production with Chat
AI Times (Korea)
Translated from Korean
-
Peruvian·Chile, re-producing power grid connection...Solar power exchange award
AI Times (Korea)
Translated from Korean
-
Copyright(c) 2014. All Rights Reserved.
AI Times (Korea)
Translated from Korean
-
Copyright (C) 2019. All Rights Reserved.
AI Times (Korea)
Translated from Korean
-
Could On-Policy Self-Distillation Revolutionize Post-Training?
Towards Data Science
Teaching foundation models new skills without breaking them is hard. A new reinforcement learning-based approach might have the power to change that. The post Could On-Policy Self-Distillation Revolutionize Post-Training? appeared first on…
-
Three things you should know about permafrost
MIT Technology Review Biotech
How much do you know about permafrost? Maybe, like me, you’ve been aware that this layer of frozen soil is thawing, that it’s important in the global puzzle of climate change … and that’s about the extent of it. The great news for both of…
-
Making the case for context-aware AI
MIT Technology Review Biotech
Enterprise AI can generate recommendations, automate tasks, and help organizations make decisions faster. But without an understanding of how a business actually operates, even powerful models can miss the constraints and relationships…
-
[AIReader's Book] Just write Postgres
AI Times (Korea)
Translated from Korean
-
AI’s hidden storage bill: What inference leaves behind
MIT Technology Review Biotech
Ask most executives to picture the cost of AI and they’ll likely see GPUs—racks of accelerators with gigawatts of power—and a waitlist for silicon. But that picture leaves something out: Data storage costs. What happens to data after a…
-
Ensemble forecasting with IBM Granite time-series models
Hugging Face Blog
-
[October 21] Why Jenson Huang directly falls on the core boss crisis
AI Times (Korea)
Translated from Korean
-
OpenAI launches the latest model in Google Cloud...
AI Times (Korea)
Translated from Korean
-
What FDE means at OpenAI
OpenAI Blog
What forward deployed engineering means at OpenAI
-
AI Uninterrupted the European government and institutional sites in malicious PHP security vulnerabilities
AI Times (Korea)
Translated from Korean
-
Unit-Aware Differentially Private Quantum Machine Learning for Clinical and Wearable Health Data
arXiv cs.AI
arXiv:2610.28968v1 Announce Type: cross Abstract: Clinical registries and wearable sensors support early health-risk assessment, yet models trained on these records may leak sensitive information when released. Differential privacy (DP)…
-
Rethinking Time Series Tokenization from a Frequency Perspective
arXiv cs.LG
arXiv:2610.28997v1 Announce Type: new Abstract: Tokenization that partitions time series into subsequences has become a foundational paradigm in modern time series modeling. However, we find that current tokenization strategies inherently…
-
Lower Bounds for Smooth Nonconvex Zeroth-Order Optimization: The Price of Determinism under Low Curvature Trace
arXiv cs.LG
arXiv:2610.27550v2 Announce Type: replace Abstract: Low curvature trace, an effective rank condition, is a common explanation for why randomized zeroth-order methods scale to high-dimensional tasks such as language model fine-tuning. Can…
-
Mixture-of-Control: State-Aware Fine-Tuning for Transformer-based Models
arXiv cs.AI
arXiv:2606.31397v2 Announce Type: replace-cross Abstract: State-based fine-tuning has emerged as a compelling alternative to weight-based adaptation for transformers, updating lightweight controls into states rather than model weights…
-
Formalizing Cooperation and Competition: An AI-Assisted Mathematical Addendum to a 2018 Planning Thesis
arXiv cs.AI
arXiv:2610.29637v1 Announce Type: cross Abstract: We present an AI-assisted mathematical addendum to Strategic Trip Planning: Striking a Balance Between Competition and Cooperation (Amar, 2018). We reconstruct selected relationships…
-
Conflicted Records: Typed Adversarial Contracts for Auditable LLM Output Quality Enforcement
arXiv cs.AI
arXiv:2610.29669v1 Announce Type: cross Abstract: LLM-generated records increasingly populate production information systems where they are treated as data assets subject to the same quality expectations as records from operational…
-
Do Tabular Foundation Models Know Physics? Contamination, Units, and the Deterministic Limit
arXiv cs.LG
arXiv:2609.02766v2 Announce Type: replace Abstract: Tabular foundation models (TFMs) learn to fill in tables the way language models fill in text --- and tables are arguably the format in which most physical measurement arrives. Did they…
-
Making AI-Assisted Claims Independently Challengeable: Publication Authority and a Protocol for Falsifiable Publication Records
arXiv cs.AI
arXiv:2609.17631v2 Announce Type: replace Abstract: AI-assisted claims can appear authoritative when evidence, analysis, human authorization, presentation, and correction history describe different states. We develop Publication Authority…
-
Energy-Based Transfer for Reinforcement Learning
arXiv cs.AI
arXiv:2506.16590v2 Announce Type: replace-cross Abstract: Reinforcement learning remains sample inefficient, particularly when agents must adapt over time to related but distinct tasks, as in continual learning settings. Efficiency can be…
-
Data-Free Finite-Volume Convolutional Neural Networks for Steady Compressible Euler Flow around a Circular Cylinder
arXiv cs.AI
arXiv:2603.01001v2 Announce Type: replace-cross Abstract: A data-free finite-volume convolutional neural network (FV-CNN) is formulated for steady Euler flow around a cylinder at eight Mach numbers, 2--15. Shared-face fluxes define…
-
SYNAPSE: Framework for Neuron Analysis and Perturbation in Sequence Encoding
arXiv cs.AI
arXiv:2603.08424v2 Announce Type: replace-cross Abstract: In recent years, Artificial Intelligence has become a powerful partner for complex tasks such as data analysis, prediction, and problem-solving, yet its lack of transparency raises…
-
Is a 9B Decision Foundation Model Needed for Symptom-to-Disease Classification? Fine-Tuning Clef-Flash Against TF-IDF Baselines
arXiv cs.AI
arXiv:2610.15029v2 Announce Type: replace-cross Abstract: Disease classification from patient-reported symptom descriptions is usually approached with lightweight discriminative models or with autoregressive large language models (LLMs)…
-
Self-organizing reinforcement learning: a learner-free swarm approaches the explore-exploit optimum
arXiv cs.AI
arXiv:2610.25113v2 Announce Type: replace-cross Abstract: The explore/exploit trade-off is the core problem of reinforcement learning (RL). We show that a minimal foraging swarm approaches its optimum - the ideal-free / Nash allocation of…
-
Our Own Jev: Typed Decisions in One Forward Pass, from an In-House Judge to Open LoRA Models
arXiv cs.CL
arXiv:2610.29052v1 Announce Type: new Abstract: A typed decision maps evidence, a question and a fixed set of options to a probability for every option, computed in one forward pass, by reading a language model's logits or a small head…
-
Accept or Repair: Preserving Passage-Supported Answers in Contrastive Decoding
arXiv cs.CL
arXiv:2610.29727v1 Announce Type: new Abstract: Large language models often answer from knowledge learned during training, even when a provided passage contradicts that knowledge. To address this problem, contrastive decoding contrasts the…
-
VeeReview: Measuring Build-Verify Pairs in Agent-Authored Pull Requests from a V-Model Perspective
arXiv cs.CL
arXiv:2610.29143v1 Announce Type: cross Abstract: Context: Reviewing a pull request requires verification evidence that addresses the proposed change. Coding agent benchmarks and studies of pull request outcomes do not directly assess this…
-
A Survey of On-Policy Distillation for Large Language Models
arXiv cs.CL
arXiv:2604.00626v5 Announce Type: replace-cross Abstract: On-Policy Distillation (OPD) trains large language models with teacher guidance on student-generated trajectories, addressing the mismatch between fixed demonstrations and the…
-
HybridResearcher: Bridging the Structured-Unstructured Boundary for Deep Research Agents
arXiv cs.CL
arXiv:2610.24899v2 Announce Type: replace-cross Abstract: While LLM agents are increasingly capable of gathering and synthesizing evidence from a single homogeneous source, such as structured tables or unstructured data, real-world…
-
Latent Bridges for Multi-Table Question Answering
arXiv cs.CL
arXiv:2606.28916v2 Announce Type: replace Abstract: We introduce GRAB, a constructor-encoder-bridge pipeline for table question answering. Our method lifts relational data into an heterogeneous graph, encodes it via message passing, and…
-
Exploiting Concavity Information in Gaussian Process Contextual Bandit Optimization
arXiv cs.LG
arXiv:2503.10836v2 Announce Type: replace-cross Abstract: The contextual bandit models sequential decision-making problems in which rewards depend on both the chosen action and observed context. In many domains such as medicine, business…
-
Bio-Inspired Auditory Processing: Coupling Parallel Resonate-and-Fire Dynamics with Ternary Recurrent Networks for Speech Recognition
arXiv cs.AI
arXiv:2610.29720v1 Announce Type: cross Abstract: Spiking Neural Networks (SNNs) promise energy-efficient inference through sparse, event-driven computation, yet remain constrained on continuous, context-dependent tasks such as automatic…
-
CALLIOPE: Private Insights into AI Use, Without Embeddings
arXiv cs.LG
arXiv:2610.28797v1 Announce Type: new Abstract: We introduce CALLIOPE, a top-down framework for discovering topic hierarchies in large corpora of unstructured data (e.g., LLM conversations), while supporting differential privacy (DP)…
-
Actually, All Learnable Distribution Classes Are Privately Learnable
arXiv cs.LG
arXiv:2610.28878v1 Announce Type: new Abstract: We show that every class of distributions that is learnable up to arbitrarily small target error in total variation distance with finitely many samples is also learnable under ($\varepsilon…
-
Copying Under Repetition: From Prefix Matching to Position Hashing
arXiv cs.CL
arXiv:2610.26722v2 Announce Type: replace-cross Abstract: Frontier LLMs can copy random text almost flawlessly, yet they often overrun or truncate long blocks of repeated characters. In repeated strings, matching local content no longer…
-
STAR-XPC: Set Transformer for Estimating X-ray Transmission in Polymer Composites
arXiv cs.LG
arXiv:2610.28709v1 Announce Type: new Abstract: Evaluating the energy transmission ratio of an X-ray beam through a polymer composite is a challenging task in radiation shielding studies. The reference solution with Monte Carlo transport…
-
ReAPO: Retry-Aware Policy Optimization for Efficient Exploration in LLM Reasoning
arXiv cs.LG
arXiv:2610.28770v1 Announce Type: new Abstract: RLVR improves language-model reasoning, but commonly relies on independent stochastic rollouts for exploration, which can repeatedly sample similar failed strategies. Prior work has seen…
-
Stationarity Constraints Improve Both Energies and Densities of Machine-Learned Exchange-Correlation Functionals
arXiv cs.AI
arXiv:2610.30022v1 Announce Type: cross Abstract: Empirical exchange-correlation functionals are fitted on fixed orbitals but used self-consistently, so better energies often come with worse densities. We show that the Brillouin residual…
-
Towards Realistic Models of Change: A Causal Perspective on Decision-Making and Dynamical Systems
arXiv cs.LG
arXiv:2610.28803v1 Announce Type: new Abstract: While statistical learning has made remarkable progress in recent decades, driven by large datasets and data-driven protocols that have transformed scientific practice, it still faces a key…
-
NV-Generate-Body: 3D Latent Diffusion Model forControllable Anatomy, Tumor, and Demographic Body Mask Synthesis
arXiv cs.CV
arXiv:2610.24153v2 Announce Type: replace Abstract: Curating large labeled CT datasets is expensive and constrained by patient privacy. We present NV-Generate-BodyMask, a 3D latent diffusion model that synthesizes whole-body CT…
-
OPRD: On-Policy Reminder Distillation for Long-Horizon Coding Agents
arXiv cs.CL
arXiv:2610.21496v2 Announce Type: replace Abstract: Reliable repository-level coding requires agents not only to generate correct code but also to preserve constraints and respond to tool feedback across long interactions. However, our…
-
A Model-Free Continuous-Time RL Algorithm from Discrete-Time Off-Policy Data with End-to-End Guarantees
arXiv cs.LG
arXiv:2610.28915v1 Announce Type: new Abstract: In this paper, we study continuous-time reinforcement learning, where the state follows an unknown stochastic differential equation but only a fixed off-policy batch of discrete-time…
-
Decision language models as affordable and rapid alternatives to LLM judges of clinical generative AI
arXiv cs.AI
arXiv:2610.28826v1 Announce Type: new Abstract: Automated judges of clinical artificial intelligence (AI) outputs are necessary, but large language model (LLM) judges are costly and slow at scale. We compared Jev, a decision language model…
-
Learning to Measure and Diagnose: A Clinical-Dictionary Radiomic Bottleneck Model for Thyroid Nodule Classification
arXiv cs.CV
arXiv:2610.29738v1 Announce Type: new Abstract: Thyroid ultrasound supports the assessment of nodule malignancy, but classification predictions often lack evidence linked to familiar clinical descriptions. Radiomics provides explicitly…
-
Mitigating Query Shortcuts in Chain-of-Thought with Context-Aware Decoding
arXiv cs.CL
arXiv:2610.29518v1 Announce Type: new Abstract: While Chain-of-Thought (CoT) prompting improves the reasoning performance of large language models (LLMs), recent studies have shown that final answers do not always depend on the generated…
-
VeriEvolve: Evaluator-First Evolution for Tool-Using Agents
arXiv cs.CL
arXiv:2610.29005v1 Announce Type: cross Abstract: A central obstacle to self-improving agents is the absence of a reliable fitness signal for realistic stateful tasks. We introduce \textsc{VeriEvolve}, an evaluator-first evolution…
-
RuleSmith: Multi-Agent LLMs for Automated Game Balancing
arXiv cs.LG
arXiv:2602.06232v2 Announce Type: replace Abstract: Game balancing is a longstanding challenge requiring repeated playtesting, expert intuition, and extensive manual tuning. We introduce RuleSmith, a framework that couples multi-agent LLM…
-
A General Topological Symmetry Index for Critical Orbits In Neural Network Loss Landscapes
arXiv cs.LG
arXiv:2610.28864v1 Announce Type: new Abstract: A systematic understanding of the global structure of neural-network loss landscapes remains challenging due to their high dimensionality. Symmetries arising from network architectures and…
-
Generalization and Scaling Laws for Mixture-of-Experts Transformers
arXiv cs.AI
arXiv:2604.09175v2 Announce Type: replace-cross Abstract: We develop a theory of generalization and scaling for Mixture-of-Experts (MoE) Transformers that cleanly separates \emph{active} per-input capacity from routing combinatorics. By…
-
SANE: Automated Scorer-Specific Metamorphic Testing for LLM Judges
arXiv cs.AI
arXiv:2610.28886v1 Announce Type: cross Abstract: LLM judges increasingly mediate benchmark evaluation, model selection, data curation, and training. Their decisions are useful only when each judge responds reliably to changes affecting…
-
ROCOCO: Robust Outline Completion of Occluded Curvilinear Objects
arXiv cs.LG
arXiv:2610.29177v1 Announce Type: cross Abstract: We propose a learning-based generative model for completing curvilinear features under occlusion by environmental obstacles. To this end, we adopt a flow matching-based approach that…
-
SSMCGM-Stream: A Streaming State-Space Model for Personalized Diabetes Management
arXiv cs.LG
arXiv:2610.28976v1 Announce Type: new Abstract: Continuous glucose monitoring supports diabetes management but requires forecasting models that are accurate, interpretable, and efficient enough for real-time deployment. We present…
-
Diffusion Posterior Sampling with Uncertain Forward Models
arXiv cs.CV
arXiv:2610.29102v1 Announce Type: new Abstract: Diffusion posterior samplers reconstruct signals by combining a pretrained diffusion prior with a measurement likelihood specified by a forward model and a noise model. Calibration and…
-
CausalAlign: Causal Alignment of Video Generation Models for Consistent Ego-Motion
arXiv cs.CV
arXiv:2610.29915v1 Announce Type: new Abstract: Diffusion-based video generation models (VGMs) can produce visually plausible videos, yet their camera ego-motion often drifts over long horizons. Camera ego-motion determines how the…
-
VoxProg: 3D Spatial Reasoning through Voxel Programming
arXiv cs.CV
arXiv:2610.30127v1 Announce Type: new Abstract: Language models' abilities to interpret tasks and generate programs offer a promising basis for 3D spatial reasoning. Yet following spatial instructions in 3D scenes requires programs to…
-
Privacy in Context Is Being Overlooked in the Age of General-Purpose LLMs
arXiv cs.LG
arXiv:2610.29925v1 Announce Type: new Abstract: As large language models (LLMs) continue to grow in scope and capability, privacy concerns are on the rise. Since users increasingly interact with a single general-purpose system across many…
-
To Manipulate or Not: Opinion Dynamics in LLM Interactions
arXiv cs.CL
arXiv:2610.28875v1 Announce Type: cross Abstract: Large language models increasingly take part in extended conversations where they can both shape and be shaped by another party's views, yet most persuasion research measures only one side…
-
Delay-Guided Attention for Quality Prediction in Multistage Manufacturing
arXiv cs.LG
arXiv:2610.29220v1 Announce Type: new Abstract: Quality prediction in multistage manufacturing is challenging when inter-stage lags vary over time, as measurements taken at the same timestamp may reflect different process states. Existing…
-
Loopholing All the Way: Language Modeling with pure Self-Conditioning
arXiv cs.LG
arXiv:2610.30068v1 Announce Type: new Abstract: Recent Diffusion Language Models (DLMs) use Self-Conditioning (SC), or Loopholing, to improve their performance. While SC is traditionally seen as an auxiliary heuristic, its empirical…
-
FlashNext-RL: A System Optimization Study for Long-Horizon Agentic RL of Qwen3.8-Flash-Next
arXiv cs.LG
arXiv:2610.28998v1 Announce Type: new Abstract: Long-horizon agentic reinforcement learning (RL) requires both efficient rollout generation and training on long sequences. However, the memory requirements and model-specific operators of…
-
HarnessPAI: An Evolving Harness for Physical AI
arXiv cs.AI
arXiv:2609.29166v2 Announce Type: replace-cross Abstract: Physical AI aims to build embodied agents that perceive the world, understand and reason about it, and decide how to act. Yet the field has focused primarily on the last component…
-
WaveCommitments: Private and Verifiable Speech-Agent Marketplaces
arXiv cs.CL
arXiv:2610.26867v2 Announce Type: replace Abstract: Modern speech systems are evolving into ecosystems of autonomous speech agents that interact, negotiate, and trade assets such as embeddings, spectrograms, or raw data. In such markets…
-
Simply Logical -- The First Three Decades
arXiv cs.AI
arXiv:2610.28912v1 Announce Type: new Abstract: This paper charts the evolution of the Prolog textbook Simply Logical -- Intelligent Reasoning by Example from print with runnable programmes on a 3.5-inch diskette published in 1994, via…
-
Approximate Solomonoff Induction: On the Robustness of Learned Universal Predictors
arXiv cs.LG
arXiv:2610.29289v1 Announce Type: new Abstract: Why large language models trained via log-loss minimization over massive and diverse datasets exhibit broad capabilities and generalizable skills is not yet fully understood. Grau-Moya et al…
-
DisasterRoute: Evidence-Guided Dual-System Routing for Visual Disaster Question Answering
arXiv cs.CV
arXiv:2610.29231v1 Announce Type: new Abstract: Visual question answering from low-altitude UAV imagery must deliver timely and accurate judgments to support disaster response under limited edge computing resources. Routine questions may…
-
Just Respect the Semantics: Image-Token Compression via Pre-LLM Semantic Variable-Resolution Tokenization and In-LLM Relevance Pruning
arXiv cs.CV
arXiv:2610.18690v2 Announce Type: replace Abstract: Image-token compression must reduce computation without hurting the semantics. We introduce Varest, guided by semantic fungibility: tokens within a coherent semantic region can be…
-
The CIDOC CRM Periodic Table: Documentation You Can Test to Support the Adoption of Scientific Research Ontologies for the Humanities
arXiv cs.AI
arXiv:2610.28650v1 Announce Type: cross Abstract: The CIDOC CRM family is not only a schema for museum collections. Its extensions formalise the apparatus through which research knowledge is produced and justified: CRMsci models scientific…
-
A General Topological Symmetry Index for Critical Orbits In Neural Network Loss Landscapes
arXiv cs.AI
arXiv:2610.28864v1 Announce Type: cross Abstract: A systematic understanding of the global structure of neural-network loss landscapes remains challenging due to their high dimensionality. Symmetries arising from network architectures and…
-
Value-Rectified Distillation for Flow-based Offline Reinforcement Learning
arXiv cs.AI
arXiv:2610.29075v1 Announce Type: cross Abstract: We aim to address the problem of Out-of-Distribution (OOD) Mode Averaging, a critical problem that emerges when distilling Flow Matching (FM) based behavior policies in offline…
-
Rethinking Agentic Model Orchestration for Time Series Forecasting
arXiv cs.LG
arXiv:2610.29504v1 Announce Type: new Abstract: Time series foundation models (TSFMs) perform unevenly across forecasting tasks. Recent work uses agentic model orchestration (AMO) to select or combine TSFMs, often outperforming individual…
-
BiMoGen: Bidirectional Motion-Text Generation via Unified Masked Discrete Diffusion
arXiv cs.CV
arXiv:2609.35407v2 Announce Type: replace Abstract: Text-to-motion generation and motion-to-text captioning are two fundamental tasks in human motion modeling, both grounded in the same underlying motion-text correspondence. Existing…
-
Telling FIBS: Probes Effectively Detect Sabotage and Catch Unverbalized Deception
arXiv cs.LG
arXiv:2610.12445v4 Announce Type: replace Abstract: Recent incidents have highlighted the challenge of monitoring LLM agents and the danger of models deceiving people. We show that white-box deception detection via probes can be scaled up…
-
Inheritance as an Admission Test for LLM Prompt Merge Operators
arXiv cs.AI
arXiv:2610.29023v1 Announce Type: new Abstract: Large language model (LLM) agents are often improved by iterating until a fixed verifier is done. Once that verifier saturates, the loop has nothing left to climb. We instead score agents…
-
Flout at Your Own Risk: LLMs Struggle with Pragmatic Cooperativity Under Epistemic Asymmetry
arXiv cs.AI
arXiv:2607.11053v2 Announce Type: replace-cross Abstract: Fruitful collaborations require effective communication of privately-held information. The increasing use of LLMs as "collaborators" raises questions about the extent to which they…
-
MCLR: Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives
arXiv cs.AI
arXiv:2603.22364v4 Announce Type: replace-cross Abstract: Diffusion models achieve strong generative performance but often rely on classifier-free guidance (CFG), an inference-time intervention that modifies the sampling trajectory. Since…
-
Kinetic theory for Transformers and the lost-in-the-middle phenomenon
arXiv cs.LG
arXiv:2605.09213v2 Announce Type: replace-cross Abstract: We study causal self-attention dynamics -- a toy model for decoder Transformers -- which we interpret as a non-exchangeable interacting particle system. Adapting cumulant expansions…
-
To Manipulate or Not: Opinion Dynamics in LLM Interactions
arXiv cs.AI
arXiv:2610.28875v1 Announce Type: new Abstract: Large language models increasingly take part in extended conversations where they can both shape and be shaped by another party's views, yet most persuasion research measures only one side of…
-
Hidden Messages Between Diffusion Language Model Agents: Causal Evidence for Latent Communication
arXiv cs.AI
arXiv:2610.28905v1 Announce Type: new Abstract: Diffusion language models form answers through a sequence of revisable continuous states, but multi-agent systems usually discard those states and make agents communicate only after decoding…
-
GenXR-Bench: Benchmarking Coding Agents for Interactive XR Development
arXiv cs.AI
arXiv:2610.29014v1 Announce Type: new Abstract: Coding agents are rapidly advancing from single-file generation to long-horizon repository-scale engineering, yet the evaluation of their capabilities to author interactive extended reality…
-
MERP: A Latent-Bottleneck Framework for Robust Multi-Modal Multi-Label Emotion Recognition
arXiv cs.AI
arXiv:2610.29430v1 Announce Type: new Abstract: Multimodal multi-label emotion recognition is important for affect-aware systems, but models trained with complete text, visual, and acoustic streams often degrade sharply when some streams…
-
RubricOpt: Self-Revision of Semantic Rubrics for Preference Optimization
arXiv cs.AI
arXiv:2610.29491v1 Announce Type: new Abstract: A model's failures can reveal gaps in the behavioral criteria guiding its training. We introduce RubricOpt, a framework that lets a policy revise these criteria within preference…
-
Brittle by Design? Multi-Scenario Reinforcement Learning for Robust Mission Coordination in Systems of Systems
arXiv cs.AI
arXiv:2610.28818v1 Announce Type: cross Abstract: As systems engineering (SE) shifts from monolithic systems to Systems of Systems, Mission Engineering (ME) has emerged as a new line of thinking in SE. Mission environments are uncertain…
-
From Reasoning to Action: Benchmarking Frontier AI Models on LIBERO-10 Robot Manipulation Tasks
arXiv cs.AI
arXiv:2610.30051v1 Announce Type: cross Abstract: Recent advances in frontier AI models have demonstrated remarkable capabilities in visual understanding, reasoning, planning, and tool use, raising the possibility of using general-purpose…
-
The End of Consensus Research?
arXiv cs.AI
arXiv:2610.30114v1 Announce Type: cross Abstract: Consensus is a fundamental problem in distributed computing and at the core of state machine replication (SMR). Despite decades of active human research, discovering practical low-latency…
-
Lean4Agent: Formal Modeling and Verification for Agent Workflow and Trajectory
arXiv cs.AI
arXiv:2606.06523v3 Announce Type: replace Abstract: Equipping Large Language Models (LLMs) to execute reliable multi-step workflows has become a central challenge in artificial intelligence. Despite recent advances in LLMs' agentic…
-
Quality-Diversity Red-Teaming: Automated Generation of High-Quality and Diverse Attackers for Large Language Models
arXiv cs.AI
arXiv:2506.07121v3 Announce Type: replace-cross Abstract: Ensuring the safety and robustness of large language models (LLMs) is a fundamental challenge and a critical prerequisite for the responsible deployment of artificial intelligence…
-
Can synthetic data reproduce real-world findings in epidemiology? A replication study using adversarial random forests
arXiv cs.AI
arXiv:2508.14936v4 Announce Type: replace-cross Abstract: Synthetic data hold substantial potential to address practical challenges in epidemiology due to restricted data access and privacy concerns. However, many current methods suffer…
-
On the Expressive Power and Limitations of Multi-Layer SSMs
arXiv cs.AI
arXiv:2604.14501v3 Announce Type: replace-cross Abstract: We study how depth, finite precision, state dimension, and chain-of-thought (CoT) affect the expressive power of multi-layer state-space models (SSMs). For the explicit-table…
-
DualView: Adaptive Local-Global Fusion for Multi-Hop Document Reranking
arXiv cs.AI
arXiv:2605.18767v2 Announce Type: replace-cross Abstract: Multi-hop question answering requires aggregating information from multiple documents, a critical capability for knowledge-intensive applications. A fundamental challenge lies in…
-
Metropolis-Scale Resilient and Trustworthy Traffic Flow Inference Using Multi-Source Data
arXiv cs.AI
arXiv:2605.25004v2 Announce Type: replace-cross Abstract: Inferring network-wide traffic states from sparse observations with high accuracy and trustworthy uncertainty quantification is essential for intelligent transportation systems, yet…
-
PaceVLA: Adaptive Temporal Action Quantization in Vision-Language-Action Models
arXiv cs.AI
arXiv:2610.25961v2 Announce Type: replace-cross Abstract: Fast and reliable manipulation is crucial for practical robot deployment. Although Vision-Language-Action (VLA) models have recently shown remarkable progress as generalist robot…
-
Uncensored Is Not Unbiased: Abliteration Removes Refusals but Not Script-Dependent Bias About Taiwan
arXiv cs.CL
arXiv:2610.29342v1 Announce Type: cross Abstract: "Abliterated" ("uncensored") builds of open-weight language models remove the activation direction that mediates refusal, and are popular partly on the assumption that a model that no…
-
Mitigating Reward Hacking in Preference Reward Models with Evolving Penalties
arXiv cs.CL
arXiv:2610.26496v2 Announce Type: replace Abstract: Reinforcement learning for language models has moved from preference reward models to verifiable rewards, which sustain much longer optimization. Preference models are prone to reward…
-
Signals and Queries: Token-Efficient Agentic Context Discovery
arXiv cs.CL
arXiv:2610.29326v1 Announce Type: new Abstract: User-facing LLM agents must identify which behavioral policies apply at each turn, but evaluating every policy with an LLM is costly. We introduce Signals and Queries (S&Q), a lightweight…
-
TNODEV: Toolbox for Neural ODE Verification
arXiv cs.LG
arXiv:2606.16567v3 Announce Type: replace-cross Abstract: Neural ordinary differential equations (neural ODE) gained attention in safety critical settings such as continuous-time controllers for cyber-physical systems and classifiers…
-
AI-Based End-to-End Automotive Design and Software Development for SDVs
arXiv cs.AI
arXiv:2610.29160v1 Announce Type: cross Abstract: This paper explores the potential of Artificial Intelligence (AI) to enable end-to-end automotive design and software development for Software-Defined Vehicles (SDVs). Based on our…
-
Why is AI-generated language so hard to understand?
arXiv cs.AI
arXiv:2610.29162v1 Announce Type: cross Abstract: There is a widespread sense that AI-generated text has recently become harder to understand. In this study, we investigate a set of psycholinguistic features that are known to correlate…
-
CAVE: A Structured Credit Assignment Approach for Fragmented Visual Evidence Reasoning
arXiv cs.AI
arXiv:2605.16416v2 Announce Type: replace-cross Abstract: Vision-Language Models (VLMs) have achieved strong performance on general multimodal reasoning, yet remain challenged in integrating nonlocal visual information to support…
-
Beyond Policy Discrepancy: Capability-Aware Multi-Teacher On-Policy Distillation
arXiv cs.LG
arXiv:2610.26950v2 Announce Type: replace Abstract: Multi-teacher on-policy distillation (MOPD) integrates domain specialists into a shared student, but determining how strongly each teacher should supervise remains challenging. Existing…
-
SAIL: A Learnable Generalisation of Spatial Autocorrelation over Graph-Structured Data
arXiv cs.CV
arXiv:2610.28806v1 Announce Type: new Abstract: In vision and scientific domains involving spatial data, a key question is whether node feature values vary independently across space or form coherent spatial structures. Classical spatial…
-
Data-Driven Glyph Positioning for Handwritten Music Synthesis
arXiv cs.CV
arXiv:2610.28899v1 Announce Type: new Abstract: Synthesised data is a common strategy to address the lack of annotated handwritten music scores in Optical Music Recognition (OMR). Smashcima is a recent synthesis framework that assembles…
-
A Novel Schur Decomposition-Based Weight Projection Method for Stable State-Space Neural-Network Architectures
arXiv cs.LG
arXiv:2605.14489v3 Announce Type: replace Abstract: Building black-box models for dynamical systems from data is a challenging problem in machine learning, especially when asymptotic stability guarantees are required. In this paper, we…