Artificial Intelligence
Artificial Intelligence news — Friday 16 October 2026
150 stories from 9 sources, every one in English. Each links to its publisher.
-
How Itaú is building generative AI for its entire customer base
OpenAI Blog
With ia.i, a conversational interface, and the IARA governance platform, Itaú is applying its AI First strategy to customers and employees — in one of Brazil’s most regulated sectors.
-
Open source and on device: the strategy behind Nvidia’s new AI models
AI Business
The new Nemotron 3.5 models show how the AI giant is focusing on reinforcing its own GPU infrastructure while supporting the open source community.
-
TypeSafe AI expands functions such as multimodalization of crystal models
AI Times (Korea)
Translated from Korean
-
MS releases 2.8 trillion parameter models prior to the release of the official release...‘The ability to reduce openAI dependence’
AI Times (Korea)
Translated from Korean
-
Google Founder “Infra, AI participates in the development of the subsequent model... Enter RSI track”
AI Times (Korea)
Translated from Korean
-
Amazon, daily conversation memory, personal AI wearable 'non' series released
AI Times (Korea)
Translated from Korean
-
TSMC, MSK’s ‘Terraf’ consultation check...
AI Times (Korea)
Translated from Korean
-
[Bulletin] KT, AI-based burial music service ‘Deobiz’ launch and others
AI Times (Korea)
Translated from Korean
-
Twelve Labs Launches AI Agent ‘Ja’ to find and edit scenes when requested by horse
AI Times (Korea)
Translated from Korean
-
Mercury, Memory Semiconductor Price Policy Powerful criticism.. . Samsung·SK hynix match
AI Times (Korea)
Translated from Korean
-
OpenAI, the voice model ‘GPT-Live-1.1’ released.. . Improvement of 6 languages including Korean
AI Times (Korea)
Translated from Korean
-
Intellectual Property Co., Ltd., listed on the 30th Co., Ltd....
AI Times (Korea)
Translated from Korean
-
OpenAI Do-Cheong Anaerobic Staff, Saudi Prince Do-Cheong Do-Cheong
AI Times (Korea)
Translated from Korean
-
Antarpic, an external evaluation organization METR allows access to ‘Employee Class’ system
AI Times (Korea)
Translated from Korean
-
Prompt: Enterprise controls can't keep up with AI agents
AI Business
Companies are giving agents more authority, but many are still struggling to track what they can access, what they're doing and how to stop them.
-
Do You Still Own the Code Your AI Wrote? I Built CodeOwnership to Measure It
Towards Data Science
I rated my confidence 4 out of 5, then couldn’t find a planted bug in 8 minutes. Here is the tool, the run, and its limits. The post Do You Still Own the Code Your AI Wrote? I Built CodeOwnership to Measure It appeared first on Towards…
-
Participated in 59 companies
AI Times (Korea)
Translated from Korean
-
Robots can't talk to each other yet -- that's a big AI bottleneck
AI Business
Coordinating robots and autonomous systems will be critical to unlocking physical AI, but it’s still unclear whether interoperability is an immediate barrier to adoption.
-
"Korean Muyang More Naturally"...OpenAI, "ChatGPT Voice" Update
AI Times (Korea)
Translated from Korean
-
"GLM-5.3, Safety Testing... .'Intense Camouflage' Possibility" Alert
AI Times (Korea)
Translated from Korean
-
I Built a Claim-Checker for Writing With Codex. It Passed a Wrong Claim.
Towards Data Science
The tool found the right passage but missed the mismatch, and the audit shows where rule-based checks break down. The post I Built a Claim-Checker for Writing With Codex. It Passed a Wrong Claim. appeared first on Towards Data Science .
-
Opening of the first Resident Artistship Fair in Gyeongsangnam-do, Republic of Korea...
AI Times (Korea)
Translated from Korean
-
Antropic, the first investor explained the IPO expectation... “Body Value 2→ 1.5”
AI Times (Korea)
Translated from Korean
-
FTC, openAI·andtropicture of the summit..."AI Safety accident·Consumer Damage Survey"
AI Times (Korea)
Translated from Korean
-
Deploying Data Science Tools: Part 1
Towards Data Science
Getting your data science code running anywhere The post Deploying Data Science Tools: Part 1 appeared first on Towards Data Science .
-
The Download: failed social media bans, and extending pets’ lives
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. Kids are outsmarting social media bans. That’s not a problem we need to solve. —Michal Luria, Aliya…
-
NVIDIA opens 120B ‘NEMRON 3.5 Super VL’
AI Times (Korea)
Translated from Korean
-
Clean hydrogen peroxide with photovoltaic...Shieldrogen, efficiency 12.3% record
AI Times (Korea)
Translated from Korean
-
2 solar grounds. . Detachable Overload 45~52%
AI Times (Korea)
Translated from Korean
-
In Europe’s solar industry, EU rebuilding supply chain... “8 million jobs in 30GW production”
AI Times (Korea)
Translated from Korean
-
I Reran My Python 3.14 JIT Benchmarks on Python 3.15
Towards Data Science
The same three workloads show a much stronger JIT, but it still doesn't help every program The post I Reran My Python 3.14 JIT Benchmarks on Python 3.15 appeared first on Towards Data Science .
-
Smile Shakes, AWS Data Computing Analytics Award...Data"Economics Computing"
AI Times (Korea)
Translated from Korean
-
Love is infrastructure
MIT Technology Review Biotech
I. Meet cute Barnabas and Mainspring grew up near at hand, about an hour apart with good traffic, but for a long time they weren’t close. They traded glimpses and occasional hellos and then went about their business. Both of them carried…
-
Can’t bear to say goodbye to your pet? Biotech’s got you covered.
MIT Technology Review Biotech
I don’t remember much about Peppy, my first and only pet cat. I can conjure the sensation of stroking her while she purred, recall how she used to get stuck in trees (just like in cartoons!), and chuckle about the time she pooed in my baby…
-
This implant can restore sight for patients with macular degeneration
MIT Technology Review Biotech
More than 5 million people around the world have an advanced form of age-related macular degeneration (AMD) called geographic atrophy, in which cells in the retinas die off. And until now, no treatment could restore their vision. The PRIMA…
-
[AIReader Books] Vector Database
AI Times (Korea)
Translated from Korean
-
Kids are outsmarting social media bans. That’s not a problem we need to solve.
MIT Technology Review Biotech
Social media bans are stacking up. In July, France joined Australia in restricting minors from accessing social media—barring children under 15 from major platforms, before the ban was blocked by the country’s top court on human rights…
-
[October 15] AI Mathematics Advancement Wavelength... From Science Research to Internet Security
AI Times (Korea)
Translated from Korean
-
Meta, ‘Muse Spark Open’ released...Powered in 2 seconds, the highest performance in the USA
AI Times (Korea)
Translated from Korean
-
Anticipatory Commonsense with Moving Objects: An ASP-Based Framework for Predictive Reasoning with Interacting Trajectories
arXiv cs.AI
arXiv:2610.20867v1 Announce Type: new Abstract: We present a general declarative framework for anticipatory reasoning about interactive motion in complex dynamic environments. Our focus is on (explainable) anticipatory space-time mental…
-
Decoding Naturalistic Speech from fMRI with Temporal Modeling and Multimodal Alignment
arXiv cs.LG
arXiv:2610.22049v1 Announce Type: new Abstract: Reconstructing what a person perceives from their brain activity is a stringent test of whether neural recordings carry meaningful information. Naturalistic speech offers a rich testbed for…
-
MeddeX: A Retrieval-Augmented LLaMA-3 System with Gemini Vision for Evidence-Backed Medical Report Analysis
arXiv cs.CL
arXiv:2610.20632v1 Announce Type: new Abstract: Modern healthcare generates increasingly complex medical reports that challenge clinical professionals in terms of time and interpretive burden. Manual review processes are slow, error-prone…
-
Does Morphological Correctness Matter for Tokenization? A Scrambled Control Study in Tagalog
arXiv cs.LG
arXiv:2610.20859v1 Announce Type: cross Abstract: Typology-aware tokenizers are motivated by the claim that segmentation which respects morpheme boundaries preserves information that frequency-driven segmentation destroys. Studies…
-
Where Is the "Self" of a Large Language Model? -- From Self-Index to Self-Reference
arXiv cs.CL
arXiv:2610.16717v2 Announce Type: replace Abstract: Does a large language model have a self, and if so, of what kind? We separate two functional levels familiar from biology: a self-index, which distinguishes one's own material from…
-
Do VLAs Really Follow Instructions? Diagnosing How-Level Instruction Following in Vision-Language-Action Models
arXiv cs.CV
arXiv:2610.21251v1 Announce Type: new Abstract: Vision-language-action (VLA) models are expected to follow instructions like their language backbones, yet prior work has focused on What-level instructions that specify the task rather than…
-
DLOSAC: Joint Spatial-Geometric Image Estimation via Delaunay-Guided Local Optimization RANSAC
arXiv cs.CV
arXiv:2610.21215v1 Announce Type: new Abstract: Reliable image geometric estimation from noisy matching points is typically addressed using RANSAC, often combined with local optimization to refine inliers and model parameters. However…
-
LVMT: Video Mask Transformer for Long-term Video Segmentation
arXiv cs.CV
arXiv:2609.34895v3 Announce Type: replace Abstract: Existing online video segmentation methods struggle to track objects in long, complex videos with long-term occlusions. We hypothesize that this limitation is caused by (i) the inability…
-
Reasoning Error from Known Fact: Step-Level Self-Consistency Group Relative Policy Optimization for LLMs
arXiv cs.CL
arXiv:2607.18915v2 Announce Type: replace Abstract: With the rapid advancement of large language models (LLMs), modern systems not only possess strong foundational capabilities and extensive knowledge, but can also solve complex problems…
-
Following Delegation Traces to Study Human Oversight in Interaction with Multimodal AI
arXiv cs.AI
arXiv:2610.21948v1 Announce Type: cross Abstract: As multimodal AI systems are becoming our everyday thought partners, the shift towards studying how users transfer control to these systems becomes especially urgent. In this paper, we…
-
Close Enough is Nowhere Near: Fidelity-Consumability Decoupling in KV Caches
arXiv cs.LG
arXiv:2610.22016v1 Announce Type: new Abstract: Multi-agent LLM systems communicate in text, which is lossy and wasteful: the sender throws away the reading it already did. A growing literature instead transmits the KV cache, the natural…
-
Right Answer, Wrong Tokens: Tail-Aware Credit Calibration for RLVR
arXiv cs.CL
arXiv:2607.07976v2 Announce Type: replace Abstract: Although reinforcement learning (RL) improves the reasoning capabilities of large language models (LLMs), GRPO-style methods can suffer from a positive-credit contamination failure mode…
-
A Bethe-Ansatz Source of Early Token Correlations: A marginal-preserving construction for parallel proposals
arXiv cs.CL
arXiv:2610.16781v2 Announce Type: replace Abstract: Parallel token proposals need a joint law, not merely accurate token marginals. We construct coordinate-Bethe-ansatz laws from the periodic XXZ chain and compare them with their…
-
ContinualSkillBench: Can LLM Agents Truly Evolve Their Capabilities?
arXiv cs.CL
arXiv:2608.03874v3 Announce Type: replace-cross Abstract: Modern agent frameworks equip large language models with external skill libraries to solve complex tasks. However, it remains unclear whether these systems can effectively evolve…
-
When Correctness Feedback Erases Evidence: Phase-Separated Completeness Auditing in Verus
arXiv cs.AI
arXiv:2610.20676v1 Announce Type: new Abstract: Correctness feedback suppresses the very evidence that specification-completeness auditing is designed to uncover: programs that satisfy a written contract but violate intended behavior. We…
-
RMBench: Memory-Dependent Robotic Manipulation Benchmark with Insights into Policy Design
arXiv cs.AI
arXiv:2603.01229v4 Announce Type: replace-cross Abstract: Robotic manipulation policies have made rapid progress in recent years, yet most existing approaches give limited consideration to memory capabilities. Consequently, they struggle…
-
Scaling Multilingual Mixture-of-Experts: How Language Coverage Changes Capacity and Compute Allocation
arXiv cs.LG
arXiv:2610.20247v2 Announce Type: replace-cross Abstract: Multilingual models share capacity across languages, so adding languages can hurt performance on existing ones. Mixture-of-experts models expand this capacity with little additional…
-
DPA-MoE: A Runtime Precision Control Framework for MoE Inference
arXiv cs.CL
arXiv:2610.21378v1 Announce Type: new Abstract: Mixture-of-Experts (MoE) models reduce computation by routing each token to a small set of experts, but each selected expert is still used at a fixed precision. We present DPA-MoE, a runtime…
-
X-Alpha: An AI Quant Researcher with Memory-Driven Self-Evolution
arXiv cs.CL
arXiv:2607.08332v3 Announce Type: replace Abstract: Human quantitative researchers mine alpha through an iterative process: they derive hypotheses from financial knowledge, implement and test them as factors, and use the resulting evidence…
-
Visual Sanad: Auditing Vision-Language Models for False Acceptance of Qur'anic Citations
arXiv cs.CV
arXiv:2610.22009v1 Announce Type: new Abstract: Vision-language models (VLMs) are increasingly used to interpret religious, educational, and culturally sensitive visual content, yet their reliability on citation verification remains…
-
TabClustPFN: A Prior-data Fitted Network for Tabular Data Clustering
arXiv cs.LG
arXiv:2601.21656v4 Announce Type: replace Abstract: Prior-data Fitted Networks (PFNs) have reframed supervised tabular learning as single-pass in-context inference without per-dataset optimization. Extending this paradigm to clustering is…
-
Adaptive Randomized Neural Networks for PDEs: Structure-Aware Feature Redistribution and Residual-Driven Block Enrichment
arXiv cs.LG
arXiv:2610.21077v1 Announce Type: cross Abstract: Randomized neural networks retain an output-linear least-squares solve, but globally sampled random features may fail to capture the spatial and scale structure of PDE solutions. We propose…
-
From Adults to Children: Benchmarking Foundation Models for Chest X-Ray Classification under Domain Shift
arXiv cs.CV
arXiv:2610.21809v1 Announce Type: new Abstract: Foundation models have shown significant potential across diverse fields and have been applied to medical imaging, but their robustness across patient populations remains unclear. In this…
-
FutureLift: Lifting Trajectory Proposals into Bidirectional Future Graphs for Motion Forecasting
arXiv cs.CV
arXiv:2610.22215v1 Announce Type: new Abstract: Accurate motion prediction is critical for safe autonomous driving, yet even strong forecasters leave residual errors in their multimodal proposals. Existing refiners compress complete…
-
ToolBottleneckBench: Measuring Tool-Agent Performance under Controlled Pressures and Fixed Harness Budgets
arXiv cs.LG
arXiv:2610.21098v1 Announce Type: cross Abstract: The success score of a tool agent reflects both its model and the budgets of the harness that runs it. We present ToolBottleneckBench, a benchmark that measures tool-agent performance under…
-
ARIA: Runtime Inspection and Amendment for Multi-Agent Systems
arXiv cs.AI
arXiv:2610.20780v1 Announce Type: new Abstract: In LLM-based multi-agent systems (MAS), a local mistake can propagate as other agents build on it, turning a correctable error into a failed task. Diagnosing completed runs or repeatedly…
-
Learning from Interaction Experience via Localized Context Distillation
arXiv cs.AI
arXiv:2610.21427v1 Announce Type: new Abstract: Learning from interaction experience is essential for enabling large language models to continually improve during deployment. Context distillation offers a promising path: privileged…
-
ReTerm: Recovery Training for Self-Improving Terminal Agents
arXiv cs.LG
arXiv:2610.18484v2 Announce Type: replace-cross Abstract: Terminal agents have shown strong abilities in complex coding tasks. However, they often struggle to recover from errors during long-running execution (e.g., command errors, faulty…
-
Pixels to Tokens: Token Space Efficient Active Learning for Low-Budget Semantic Segmentation
arXiv cs.CV
arXiv:2610.20881v1 Announce Type: new Abstract: While foundation models and vision transformers (ViTs) have redefined visual representations, current active learning (AL) strategies for these backbones largely operate at the patch or…
-
DefoliationCAM: An Edge-AI System for Real-Time Insect Defoliation Severity Assessment in Grapevines
arXiv cs.CV
arXiv:2610.20879v1 Announce Type: new Abstract: Insect-induced defoliation threatens grapevine health and yield, yet severity assessment relies on subjective manual scouting. We developed DefoliationCAM (D-CAM), a solar-powered edge AI…
-
Decision-FLIP: Budget-Free Structured Pruning for Gaussian Occupancy Models
arXiv cs.CV
arXiv:2610.20917v1 Announce Type: new Abstract: Occupancy predictors rely on large persistent banks of Gaussians or learned queries whose size is fixed by hand. Common pruning proxies, such as opacity, activation frequency, or parameter…
-
Trajectory Prediction Method for person under WSN with Hierarchical Envelope Adversarial Domain Adaptation
arXiv cs.CV
arXiv:2610.21132v1 Announce Type: new Abstract: Pedestrian trajectory prediction holds significant applications across autonomous driving, intelligent surveillance, and smart cities. However, the complexity and unpredictability of…
-
The Dormant Spiking Neuron: A State-Driven Mechanism for Efficient Spiking Neural Networks
arXiv cs.CV
arXiv:2610.21748v1 Announce Type: new Abstract: Spiking neural networks (SNNs) offer a compelling pathway toward energy-efficient artificial intelligence, fundamentally anchored in presynaptic input-driven sparsity. However, a substantial…
-
Learning to Think and Explore in 3D World
arXiv cs.CV
arXiv:2610.21783v1 Announce Type: new Abstract: We study how vision-language models (VLMs) can learn to think and actively explore in 3D space. Instead of providing multi-view images, videos, or 3D representations to the model, we preserve…
-
Teaching DINOv3 About Partial 3D Geometry: A Self-Supervised Geometry-Aware Approach
arXiv cs.CV
arXiv:2610.21802v1 Announce Type: new Abstract: Partial shape matching is a crucial yet underexplored problem in 3D vision, with significant relevance to real-world scenarios where shapes are often only partially observed. Existing feature…
-
VesperMUSE: Multi-Modal Fusion Vision Transformer for Image Semantic Segmentation Under Diverse Light Conditions
arXiv cs.CV
arXiv:2610.21990v1 Announce Type: new Abstract: Semantic segmentation is critical for the reliable navigation and scene understanding of autonomous systems, yet its performance degrades severely in low-light environments due to noise…
-
Data-Efficient ConvNeXt Pretraining for High-Resolution Sentinel-1 GRD SAR Representation Learning
arXiv cs.CV
arXiv:2610.21993v1 Announce Type: new Abstract: Labeled Sentinel-1 Ground Range Detected (S1GRD) datasets remain scarce, limiting supervised deep learning for land-surface remote sensing, while self-supervised pretraining for this domain…
-
Pathology foundation models for training-free case retrieval in salivary gland tumors
arXiv cs.CV
arXiv:2610.22087v1 Announce Type: new Abstract: In pathology, pattern recognition relies on a frame of reference built through extensive training and experience, which may still fall short when encountering rare cancer subtypes. Archives…
-
FALCON: FPGA-Accelerated Latency-Certifiable Occupancy Networks for Camera-Based Autonomous Driving
arXiv cs.CV
arXiv:2610.22143v1 Announce Type: new Abstract: A perception system that must brake for a pedestrian does not need a low average latency. It must deliver an answer before the deadline on every frame, the worst case included. Commodity edge…
-
Beyond Pairwise Repulsion: Collective Diversity via Determinant Maximization in Diffusion Sampling
arXiv cs.CV
arXiv:2610.22267v1 Announce Type: new Abstract: Strong guidance in text-to-image (T2I) diffusion models concentrates sampling toward high-density modes of the conditional distribution, collapsing independently sampled noise vectors onto…
-
MJev-Spatial: System One Spatial Intelligence via Native Attention Matching
arXiv cs.CV
arXiv:2610.22276v1 Announce Type: new Abstract: Spatial intelligence lets an agent perceive and act in the physical world, and many of its judgments must be made at interaction rates: which spatial relation holds, how far an object is, or…
-
ARC-SORT: Adaptive, Robust, Compensated Multi-Object Tracking with Interacting Multiple Models
arXiv cs.CV
arXiv:2610.22306v1 Announce Type: new Abstract: Multi-object tracking in sports broadcasts is hard for two reasons: athletes move fast and change direction abruptly, and the camera pans and zooms at the same time. We present ARC-SORT…
-
LICOA: Learning Instruction-Conditioned Optical Actions for Mobile Spatial Augmented Reality
arXiv cs.CV
arXiv:2610.21282v1 Announce Type: cross Abstract: Spatial augmented reality (SAR) augments physical environments with projected content, but most systems assume that the projector pose, target region, or motion path has been fixed or…
-
AI-Driven Disaster Response and Displacement Monitoring
arXiv cs.CV
arXiv:2610.21482v1 Announce Type: cross Abstract: ARTIFICIAL INTELLIGENCE (AI) has demonstrated its potential across a wide range of humanitarian tasks, offering scalable and data-driven solutions to some of the world's most pressing…
-
OS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents
arXiv cs.CV
arXiv:2610.22234v1 Announce Type: cross Abstract: Real desktop automation is neither purely visual nor purely programmatic. Agents must understand user-facing interfaces while efficiently manipulating files, scripts, and system state…
-
UniFGVC: Universal Training-Free Few-Shot Fine-Grained Vision Classification via Attribute-Aware Multimodal Retrieval
arXiv cs.CV
arXiv:2508.04136v4 Announce Type: replace Abstract: Few-shot fine-grained visual classification (FGVC) aims to discriminate between subtly different categories with very limited data. While Multimodal Large Language Models (MLLM) exhibit…
-
SceneReVis: A Self-Reflective Vision-Grounded Framework for 3D Indoor Scene Synthesis via Multi-Turn RL
arXiv cs.CV
arXiv:2602.09432v2 Announce Type: replace Abstract: Current one-pass 3D scene synthesis methods often suffer from spatial hallucinations, such as collisions, due to a lack of deliberative reasoning. To bridge this gap, we introduce…
-
Thinking with Images as Continuous Actions: Numerical Visual Chain-of-Thought
arXiv cs.CV
arXiv:2602.23959v2 Announce Type: replace Abstract: Recent MLLMs increasingly rely on visual chain-of-thought to perform region-grounded reasoning over images. However, existing approaches ground regions via either textified coordinates…
-
Out-of-Distribution Object Detection in Street Scenes via Synthetic Outlier Exposure and Transfer Learning
arXiv cs.CV
arXiv:2603.16122v2 Announce Type: replace Abstract: Out-of-distribution (OOD) objects are often silently overlooked by object detectors, which produce no detection for them, leaving downstream modules with nothing to act on and no…
-
HIMMEL: Hierarchical Interleaved Multi-stream Motion Encoding for Long Video Understanding
arXiv cs.CV
arXiv:2605.08158v3 Announce Type: replace Abstract: Long video understanding with multimodal language models suffers from three compounding bottlenecks: heavy decode cost to obtain dense RGB frames, token growth with frame count, and weak…
-
LCPNet: Latent Consistent Proximal Unfolding Network for Infrared Small Target Detection
arXiv cs.CV
arXiv:2607.04603v3 Announce Type: replace Abstract: Infrared small target detection (IRSTD) aims to identify long distance small targets from complex infrared backgrounds, and is a fundamental task in remote sensing. Deep learning methods…
-
LiFR v2: Completion-Augmented Event Propagation for High-Rate Dense Prediction
arXiv cs.CV
arXiv:2609.25803v3 Announce Type: replace Abstract: High-rate dense perception in dynamic environments is limited by the low update rate of RGB cameras, as rapid scene changes can occur between frames. Event cameras offer temporally dense…
-
Meaningful Moments: A Corpus of VLM-Derived Temporal Importance Labels for Video Action Recognition
arXiv cs.CV
arXiv:2610.16009v2 Announce Type: replace Abstract: Not every moment in a video is equally useful for recognizing an action. A few decisive moments often carry the evidence: the instant an object moves, the peak of a dive, or one object…
-
TwinPrune: Density-Aware Two-Phase Token Pruning for Vision-Language Models
arXiv cs.CV
arXiv:2610.17530v2 Announce Type: replace Abstract: We propose a unified spatial-density signal to score token importance for both inference phases: bipartite merge for prefill and KV eviction for decode, yielding measurable gains in each…
-
BOLT: Universal Classifier Training via a Bayes-Error Upper Bound
arXiv cs.CV
arXiv:2501.07754v2 Announce Type: replace-cross Abstract: This work introduces a novel approach to compute a universal upper bound on the Bayes error rate (BER) from classifier-output samples, which serves as a Bayes-error-related…
-
PanoDiff-SR: Synthesizing Dental Panoramic Radiographs using Diffusion and Super-resolution
arXiv cs.CV
arXiv:2507.09227v2 Announce Type: replace-cross Abstract: Interest in generating realistic synthetic medical images has grown in recent years. Such datasets can mitigate the scarcity of public data for artificial intelligence research and…
-
Real-time Chatbot Romance Scam Detection in Stage Classification
arXiv cs.CL
arXiv:2610.20631v1 Announce Type: new Abstract: Romance scam has become a serious problem via social media, and the improvement of artificial intelligence enable to fake identify easily. In the study, a chatbot has developed to detect…
-
Jev-as-a-Judge for RAG Claim Verification: Confidence-Gated Cascading Between a Non-Generative Verifier and a Frontier LLM
arXiv cs.CL
arXiv:2610.20634v1 Announce Type: new Abstract: Retrieval-augmented generation asks one question at volume: does this evidence support this claim? The answer is one bit, yet most pipelines obtain it from a text generator and pay generation…
-
Wolof-Adapted Industry-Ready Open-Weight Language Models
arXiv cs.CL
arXiv:2610.20824v1 Announce Type: new Abstract: Wolof is spoken by 23M people, mostly in Senegal, but remains weakly supported by open-weight language models, while API models raise cost, governance, and sovereignty constraints for local…
-
System1-Arena: Which Decisions Can Be Left to System 1?
arXiv cs.CL
arXiv:2610.21244v1 Announce Type: new Abstract: Large language models (LLMs) often make discrete decisions, such as flagging phishing or escalating to a human. System-1 models, such as Jev and the OpenAI Decisions API, return probabilities…
-
SITAR: A Situational Distortion Metric for Automatic Speech Recognition Evaluation
arXiv cs.CL
arXiv:2610.21440v1 Announce Type: new Abstract: In automatic speech recognition (ASR), transcription errors with similar lexical or semantic discrepancies can have markedly different consequences for the situation conveyed by speech…
-
Dynamic ordering in sequence generation in Transformers: experiments on low-resource Tagging scenarios
arXiv cs.CL
arXiv:2610.21862v1 Announce Type: new Abstract: This paper introduces TagInsert, a Transformer-based model that challenges the common left-to-right (L2R) generation paradigm in Transformers, for sequence labelling tasks where dependencies…
-
EMO: A Closed-Loop Test-Time Computation Framework via Alternating Context Optimization and Logits Steering
arXiv cs.CL
arXiv:2610.22160v1 Announce Type: new Abstract: Test-time context optimization adapts frozen language models through textual feedback, but ordinary conditional generation provides no separate control over how strongly to reinforce the…
-
Love for Believable AI: Artificial Partiality and Relationship Persistence as an Engineerable Stance
arXiv cs.CL
arXiv:2610.09895v1 Announce Type: cross Abstract: We study a behavioral mechanism for artificial partiality in conversational agents. The paper reports no human-subjects data and makes no claim about attachment, trust, perceived mind, or…
-
CogDrift-4D: Hierarchical Multimodal Modeling of Temporal Behavioral Decay for MCI Detection
arXiv cs.CL
arXiv:2610.20877v1 Announce Type: cross Abstract: Early detection of mild cognitive impairment remains a critical bottleneck in dementia care, demanding scalable tools capable of capturing subtle behavioral decay. To capture the subtle…
-
Athanor: An Agent-Native System for Autonomous Post-Training Research
arXiv cs.CL
arXiv:2610.21055v1 Announce Type: cross Abstract: Recursive self-improvement, in which AI systems build stronger AI systems, is one promising route to superintelligence. We study a weaker form that can be measured today: frontier agents…
-
PocketTSLM: Deploying an On-Device Soft-Prompt Time-Series Language Model for Privacy-Preserving Smartphone Reasoning
arXiv cs.CL
arXiv:2610.21308v1 Announce Type: cross Abstract: Consumer phones and wearables increasingly collect physiological time-series data, yet most mobile systems treat these streams as summaries or text rather than as temporal evidence. This…
-
Human Interventions as Labels: Localizing, Steering, and (Sometimes) Editing the Computations Behind LLM Agent Failures
arXiv cs.CL
arXiv:2610.21708v1 Announce Type: cross Abstract: When an LLM agent misbehaves, a human steps in, and that intervention is usually discarded or replayed as training data. We treat it instead as a measurement: a label pointing at the…
-
Documenting Trofinetide Treatment Phenotypes in Rett Syndrome with a Dual-LLM Consensus Pipeline
arXiv cs.CL
arXiv:2610.22062v1 Announce Type: cross Abstract: Objective: Trofinetide, approved in 2023 as the first Rett syndrome treatment, lacks scalable real-world monitoring because chart review is labor-intensive. We developed a dual large…
-
From Documents to Spans: Span-Centric Learning for LLM-based Explainable Medical Coding
arXiv cs.CL
arXiv:2603.15270v3 Announce Type: replace Abstract: Medical coding, which assigns standardized ICD codes to clinical documents, is a fundamental yet challenging task in modern healthcare. It requires extensive knowledge of a large code…
-
NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness
arXiv cs.CL
arXiv:2609.08183v2 Announce Type: replace Abstract: Recursive self-improvement (RSI) requires a concrete mechanism through which an AI system observes its capabilities and converts that evidence into the next round of learning. We present…
-
Analyzing performance of Retrieval Augmented Generation models for Question Answering Tasks
arXiv cs.CL
arXiv:2610.20419v2 Announce Type: replace Abstract: Retrieval-Augmented Generation (RAG) models have become almost a cornerstone of open-domain question answering tasks; in which the quality of answers generated in the process, greatly…
-
BenchPreS: A Benchmark for Context-Aware Personalized Preference Selectivity of Persistent-Memory LLMs
arXiv cs.CL
arXiv:2603.16557v3 Announce Type: replace-cross Abstract: Large language models (LLMs) increasingly store user preferences in persistent memory to support personalization across interactions. However, in third-party communication settings…
-
Spectral Indistinguishability and Learning-Curve Identifiability in Quadratic Networks
arXiv cs.LG
arXiv:2610.20678v1 Announce Type: new Abstract: We study what spectral summaries and scalar learning curves reveal about centered quadratic networks. We construct analytic positive input densities with exactly Gaussian one- and…
-
On-Policy Residual Learning for Diffusion Language Models
arXiv cs.LG
arXiv:2610.20687v1 Announce Type: new Abstract: Residual context is key to unlocking the speed-quality Pareto frontier of diffusion language models (DLMs) by carrying forward predictive distributions across denoising steps. Yet turning…
-
Clinically Interpretable Biological Age Estimation Using CatBoost and SHAP with Continuous and Categorical Features from NHANES
arXiv cs.LG
arXiv:2610.20868v1 Announce Type: new Abstract: Biological age (BA) captures inter-individual variation in physiological aging beyond chronological age (CA). Although machine-learning models have advanced BA estimation, many rely primarily…
-
Bellman-Consistent Trajectory RL for LLM Arithmetic Reasoning
arXiv cs.LG
arXiv:2610.21060v1 Announce Type: new Abstract: To address sparse terminal rewards in LLM arithmetic reasoning, we propose a Bellman-consistent trajectory RL method. Our framework introduces Trajectory Bellman Residual Minimization (TBRM)…
-
Predicting Live Birth After Fresh Embryo Transfer in IVF/ICSI: Differential ICM/Trophectoderm Morphology Encoding and a Hybrid Attention--Logistic Regression Ensemble
arXiv cs.LG
arXiv:2610.21134v1 Announce Type: new Abstract: Predicting live birth before fresh embryo transfer could support shared decision-making in IVF/ICSI, but routine clinical data are small, heterogeneous, and contain pathway-dependent…
-
Ranking and Partitioning of Attention Heads for Sparse and Continual Reinforcement Learning Fine-tuning
arXiv cs.LG
arXiv:2610.21336v1 Announce Type: new Abstract: We study sparse and continual reinforcement learning fine-tuning of language models through attention-head ranking and partitioning. For sparse learning, we rank attention heads by gradient…
-
PocketTSLM: Deploying an On-Device Soft-Prompt Time-Series Language Model for Privacy-Preserving Smartphone Reasoning
arXiv cs.LG
arXiv:2610.21308v1 Announce Type: new Abstract: Consumer phones and wearables increasingly collect physiological time-series data, yet most mobile systems treat these streams as summaries or text rather than as temporal evidence. This…
-
Spectrum Is Not Enough: A Random-Matrix Account of Data-Free Preconditioning for Differentially Private Learning
arXiv cs.LG
arXiv:2610.21312v1 Announce Type: new Abstract: Data-free preconditioning builds the Kronecker factors of KFAC from synthetic probes instead of private data. DP-KFC validates a probe by matching the eigenvalue decay of the private factors…
-
Memorilla: Latent Semantic Memory for LLMs
arXiv cs.LG
arXiv:2610.21375v1 Announce Type: new Abstract: Memory enables large language models (LLMs) to use information beyond their parameters, such as past user conversations, long documents, or an agent's earlier actions. For a frozen LLM…
-
Each Reward to Its Relevant Tokens: Reward-Specific Token-Level Credit Assignment for Multi-Reward RL
arXiv cs.LG
arXiv:2610.21439v1 Announce Type: new Abstract: Language models are increasingly trained against several verifiable rewards at once, such as the correctness and the format of a tool call, or the individual constraints of a complex…
-
Federated Art Chase: Privacy Leakage and Data Poisoning in Federated Fine-Tuning
arXiv cs.LG
arXiv:2610.21547v1 Announce Type: new Abstract: Federated learning (FL) is increasingly used to adapt a pretrained, centrally fine-tuned model to data that remain on user devices, beyond training a model from scratch. In this report…
-
Transfer Learning in Bandits from Unknown Adaptive Policies: Problem-Dependent Lower and Upper Bounds
arXiv cs.LG
arXiv:2610.21582v1 Announce Type: new Abstract: Transfer learning in stochastic multi-armed bandits is typically studied assuming that prior data forms a fixed offline dataset or is generated by a known source. In practice, historical…
-
Local Temporal Contrasts on Human Motion Data Induce a Latent Geometry
arXiv cs.LG
arXiv:2610.21630v1 Announce Type: new Abstract: Time series such as recordings of human motion data often encode structure across multiple temporal scales, and when a latent state determining this sequence evolves, this structure needs to…
-
Human Interventions as Labels: Localizing, Steering, and (Sometimes) Editing the Computations Behind LLM Agent Failures
arXiv cs.LG
arXiv:2610.21708v1 Announce Type: new Abstract: When an LLM agent misbehaves, a human steps in, and that intervention is usually discarded or replayed as training data. We treat it instead as a measurement: a label pointing at the internal…
-
Online Quantile Omniprediction for Proper Losses
arXiv cs.LG
arXiv:2610.21342v1 Announce Type: new Abstract: The task of constructing a predictor that is simultaneously accurate under multiple evaluation metrics has been popularized recently under the name omniprediction. Prior work in this area has…
-
The Lagrangian Structure of Cost-Penalized Active Feature Acquisition
arXiv cs.LG
arXiv:2610.21919v1 Announce Type: new Abstract: Active Feature Acquisition (AFA) sequentially selects features to maximize predictive accuracy under an acquisition budget. A natural heuristic greedily maximizes Expected Information Gain…
-
OS-Helix: Unifying GUI Interaction and Code Execution in Native Computer-Using Agents
arXiv cs.LG
arXiv:2610.22234v1 Announce Type: new Abstract: Real desktop automation is neither purely visual nor purely programmatic. Agents must understand user-facing interfaces while efficiently manipulating files, scripts, and system state…
-
BD-HazardVLM: Probing Vision-Language Models for Latent Defensive-Driving Hazards in Bangladesh Road Scenes
arXiv cs.LG
arXiv:2610.20627v1 Announce Type: cross Abstract: Vision-language models are increasingly used for driving-scene understanding, yet their defensive reasoning under weakly structured Global South traffic remains unclear. We introduce…
-
Feasibility Ceilings Before Learning: Sizing and Timing Bounds for Drone-Mesh Interception
arXiv cs.LG
arXiv:2610.20823v1 Announce Type: cross Abstract: A mesh of interceptors defending a fixed site must spread over the predicted arrival point of an approaching object. This is a recurring task in multi-agent robotics: N agents, each…
-
Pixels to Tokens: Token Space Efficient Active Learning for Low-Budget Semantic Segmentation
arXiv cs.LG
arXiv:2610.20881v1 Announce Type: cross Abstract: While foundation models and vision transformers (ViTs) have redefined visual representations, current active learning (AL) strategies for these backbones largely operate at the patch or…
-
Cross-subject decoding of imagined speech from scalp EEG: a leakage-controlled evaluation on three public corpora
arXiv cs.LG
arXiv:2610.20954v1 Announce Type: cross Abstract: Objective. A decoder of imagined speech that works for a new user without calibration would make a non-invasive speech interface practical. We tested whether imagined-speech content…
-
Two-Stage Score Estimation for Group-Specific Diffusion Models
arXiv cs.LG
arXiv:2610.20979v1 Announce Type: cross Abstract: Generative modeling across related distributions raises a question: how can shared structure support accurate generation for a target distribution with limited observations? We develop a…
-
Athanor: An Agent-Native System for Autonomous Post-Training Research
arXiv cs.LG
arXiv:2610.21055v1 Announce Type: cross Abstract: Recursive self-improvement, in which AI systems build stronger AI systems, is one promising route to superintelligence. We study a weaker form that can be measured today: frontier agents…
-
Pathwise Shrinkage and Probabilistic Calibration of the Multiplicative Half-Cauchy Process
arXiv cs.LG
arXiv:2610.21136v1 Announce Type: cross Abstract: We characterize the multiplicative half-Cauchy process through its log-scale random-walk representation, establishing shrinkage, expansion, and critical regimes, almost-sure summability…
-
Diffusion Language Models as Open Quantum Systems
arXiv cs.CL
arXiv:2610.21672v1 Announce Type: cross Abstract: Diffusion language models generate text by reversing a classical diffusion process over token sequences. We show that their continuous-time forward and reverse dynamics can be embedded in…
-
Minimal Witness Reinforcement Learning
arXiv cs.CL
arXiv:2610.07226v2 Announce Type: replace-cross Abstract: ``What are the irreducible conditions that are sufficient to produce an outcome?'' is one of the most common questions that recur across computation and science. Its answers, the…
-
SNOAR: Self-training with Noise filtering, Observation adding, and Remixing for Speech Denoising Without Clean References
arXiv cs.LG
arXiv:2610.20863v1 Announce Type: cross Abstract: Unsupervised speech denoising methods such as MixIT and noisy-target training can be trained directly on noisy recordings, avoiding the sim-to-real gap of synthetic training data, but…
-
SelectPFN: Amortized Bayesian Variable Selection for Sparse Nonlinear Regression with Prior-Data Fitted Networks
arXiv cs.LG
arXiv:2610.21950v1 Announce Type: cross Abstract: Tabular foundation models based on prior-data fitted networks (PFNs) predict on a new data set in a single forward pass, but they cannot report which inputs their predictions depend on. We…
-
Whole-Body Dexterity
arXiv cs.LG
arXiv:2610.22184v1 Announce Type: cross Abstract: Current humanoid dexterous manipulation remains largely confined to end-effectors and simple grasps, falling short of humans' ability to engage the entire body by bracing and sliding bulky…
-
Shortcut Sensitivity in Agent-Safety Evaluations
arXiv cs.LG
arXiv:2610.22243v1 Announce Type: cross Abstract: AI agents take actions through tools. Some actions are sensitive, so guards that supervise them must themselves be evaluated reliably. Yet the data used to evaluate a guard can include cues…
-
The impact of temporal resampling on the safety of clinical reinforcement learning: A simulated trial
arXiv cs.LG
arXiv:2602.06603v4 Announce Type: replace Abstract: Reinforcement learning (RL) is a type of artificial intelligence for making optimal choices. In healthcare, researchers generally use offline RL (ORL), where models are trained and…
-
3DTCR: Vortex-following generative reconstruction improves tropical cyclone intensity prediction from global AI weather forecasts
arXiv cs.LG
arXiv:2603.13049v3 Announce Type: replace Abstract: Global numerical weather prediction (NWP) and artificial intelligence (AI) models often underestimate tropical cyclone (TC) intensity, partly because coarse grids underresolve fine-scale…
-
Detectability without Interference: Efficient Canary Design for Privacy Auditing in One Run
arXiv cs.LG
arXiv:2605.27292v3 Announce Type: replace Abstract: Privacy auditing aims to empirically assess privacy leakage in machine learning models using membership inference attacks (MIAs), and to derive lower bounds on differential privacy (DP)…
-
When to Align, When to Predict: A Phase Diagram for Multimodal Learning
arXiv cs.LG
arXiv:2606.11190v4 Announce Type: replace Abstract: Cross-modal alignment (CA) and cross-modal prediction (CP) are the dominant paradigms for multimodal representation learning, yet there is no systematic understanding of when each…
-
Structure-Preserving Correction Learning for Sparse Bayesian Inference in Brain Source Imaging
arXiv cs.LG
arXiv:2606.07196v2 Announce Type: replace Abstract: Classical sparse Type-II Bayesian methods for M/EEG brain imaging support joint estimation of source and noise hyperparameters, but rely on fixed iterative update rules. Although these…
-
Multi-Modal, Multi-Temporal Geospatial Foundation Models Improve Global Above-Ground Biomass Estimation
arXiv cs.LG
arXiv:2608.04792v2 Announce Type: replace Abstract: Accurate estimation of above-ground biomass (AGB) from satellite imagery is essential for the large-scale monitoring of carbon stocks, yet it remains a challenging regression task at…
-
Blockwise Stabilized Adaptive Cubic Regularization with Subsolvers via Recurrence
arXiv cs.LG
arXiv:2608.22129v3 Announce Type: replace Abstract: Overparameterized networks often have ill-conditioned, saddle rich loss landscapes. They are usually trained with first-order methods that struggle on those landscapes, while many…
-
KV-streams for Efficient Compaction in Agentic Reinforcement Learning
arXiv cs.LG
arXiv:2609.35750v3 Announce Type: replace Abstract: Scaling the horizon of agentic LLMs is bottlenecked by the need to fit ever longer context traces in GPU memory. Context compaction has been the most popular mechanism to alleviate this…