Artificial Intelligence
Artificial Intelligence news — Wednesday 11 November 2026
150 stories from 9 sources, every one in English. Each links to its publisher.
-
[Nov. 11] 5 investment logics with 42-minute video
Antropicy, leading the IPO, revealed a 42-minute video to acquire investors. As a member of the Lego Amoday CEO, we will explain why we need to be recognized for more than one trillion corporate value, and we will highlight AI safety, and…
-
OpenAI’s Businesses in ChatGPT targets enterprise trust
The product competes with Google Business Profile and enables OpenAI to target a new vertical market: ecommerce.
-
With IPO looming, Anthropic launches new AI models, workplace tools
The generative vendor released new capabilities targeting enterprise work and significantly updated models as it prepares for the historic public offering.
-
Using mathematics to validate network configurations
Corporations, banks, hospitals, and academic institutions all rely on wide-area networks (WANs) to connect their local networks across geographic regions. Within a WAN, however, it’s often necessary to isolate different computational…
-
What to know about this year’s UN climate talks
It’s that time of year again: Thousands of delegates, lobbyists, and journalists are gathered for the annual UN climate talks, taking place this time in Antalya, Turkey. COP31 officially kicked off on November 8, marking the beginning of…
-
“Are You Sure?” Flipped Qwen2.5-1.5B’s Right Answer 2 Times in 3
Three small open models, 150 MMLU questions each, five pushback styles: “That’s wrong” did more damage, and naming a wrong option was close to fatal. The post “Are You Sure?” Flipped Qwen2.5-1.5B’s Right Answer 2 Times in 3 appeared first…
-
AlphaGenome 2: Deciphering how cells read DNA to better understand disease
Predicting how DNA sequences and their variations impact specific human cell types and disease.
-
AI Assistant vs. Copilot vs. Agent, and When to Use Each
A palace-hotel wedding week shows where assistants, copilots, agents, and orchestration fit, and why Gartner expects 40%+ of agentic AI projects to be cancelled. The post AI Assistant vs. Copilot vs. Agent, and When to Use Each appeared…
-
How to Use Apple's fm CLI to Run On-Device LLMs From Python
A Sate Padang vision test and a Malaysia capital question show why the free, private model needs grounding and validation before it leaves prototyping. The post How to Use Apple's fm CLI to Run On-Device LLMs From Python appeared first on…
-
The Download: preserving minds and weighing open AI risks
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. Six wild ideas for perpetuating your mind As we age, we forget most of what we’ve experienced, yet…
-
A Jev of All Trades: Build and Train a Jev-style Decision Model From Scratch
Trained on six PathMNIST tissues with a small KEEP adapter, it handled added labels without retraining, though a linear classifier hit 88% on familiar ones. The post A Jev of All Trades: Build and Train a Jev-style Decision Model From…
-
Doorshots, Chinese Nvidia Latest AI Chips For Rent... Check the Financial Documents
China MoonshotAI was found the financial contract documentation that was trying to rent the latest AI chip "B300" in Nvidia, the Chinese export regulator of the United States. The doorshot has been deemed related transactions, and since…
-
Suncheon Sunrider, Vulnerable Floor, 30th Generation Coloring Play Program Operation
A visit-type coloring play program that combines the support and cognitive activities for vulnerabilities and eagles. Since September, Suncheon Hwang-myeon has been working on ‘Tricolor Play Program’ for depression and dementia prevention…
-
Google, ‘Jeminai Robotics 2.5’ Public...Factory Job Success Rate 96%
Google released a new physical AI model, which has a high level of work success rate of the humanoid robot from a real car factory. Improved hands-on and work speeds than traditional models, and also utilized the latest AI models to robot…
-
"Fable 5.2" Prompt Guide Revealed..."Change 5 times increase
Andropic recalls the latest model 'Call Payable 5.2'' prompt writing and rechecking instructions used for existing models. In particular, how do you express the ownership of the code, it appears that AI's work is largely different…
-
"Jiminai" will show you my information at a glance... LOGIN JOIN
Google introduced the ‘Memory summary’ feature that shows information about users at a glance. Users can check their job and interests, lifestyle habits, etc. that AI holds and correct or delete them. "ChatGPT" and "Clipload" will enhance…
-
Headquarter, Headquarter, Headquarter
LG Chemistry is a management policy related to the review of the establishment of semiconductor material production facilities, such as location and permit. Yeosu said that LG Chemistry is known to review the IPA production facility in the…
-
Buying 8075 tons of public rain... Starting from 12th
Naju bought 8075 tons of public rainbow in 2026. Dry light purchase starts from 12 and is paid for KRW 60,000 for mid-range stocks per 40kg. In the end of December, the company began to buy dry light from the 12th to the end of December to…
-
Meta, AI agent ‘Muse’ expands to Apple Watch and WearOS
Meta expands the consumer AI agent ‘Muse’ into Apple Watch and Google’s WearOS smartwatch. After launching the last September, users can easily use AI agents on their smartphones and tablets to wearable devices. Meta has announced the Muz…
-
TSMC invests up to 1.3 trillion KRW in its own chip development, openAI·andtropicy
TSMC, the world’s largest Foundry company, has invested up to $10 billion (about 1 trillion) in openAI and eruption in its own AI semiconductor development. In order to reduce the dependence on external semiconductor companies such as…
-
Google, AI-encrypted weaknesses have been detected and validating signals security...650 million lines
Google uses AI to find the weakness of the RSA password, this time proved that the real encryption code of the messenger signal meets the core security conditions. OpenAI and Andropic are expanding AI utilization range from Google to…
-
$350 million IPO smoke review...AI data center investment anxiety
Nscale, the UK AI cloud company Nscale, invested in Nvidia, has reviewed the implementation of an IPO that aims to be an enterprise value of $350 billion (about KRW 51 trillion). AI data center companies are likely to have a problem with…
-
[Bulletin] Fast Campus·Environmental Development AI Agent Winner, etc.
■ Done Company (CEO Shinhae-dong·Kim Donghyuk)’s practical education brand Fast Campus was held at ‘Electronic AI Hackathon’ in collaboration with Nvidia. The final winner took place at the ‘Pladegate’ team, consisting of a new drug…
-
SoftBank, AI development equipment 5 trillion lost on order...Earth 9 trillion
Softbank has been pre-ordered with the development equipment required for AI data center in the United States. It has been lost about 5 trillion won. We have paid a huge amount of prepayments due to lack of power equipment, but because we…
-
OpenAI, Odeshi acquired and developed travel AI service
OpenAI has acquired an AI agent startup Odessia, a travel specialist. Francis Davidson, founded the Startup Hander, similar to Airbnb, is also a prospective approach to developing AI services for travel. Davidson said, “I’m going to help…
-
Google, an ad revenue deteriorating on AI search... reviewing the company acquisition
Google’s AI search increases the growth of the key revenue source search ads, and we’ve set up a response book. While some advertisers are actually starting to move advertisements to another platform, they are known to be reviewed by the…
-
OpenAI introduces corporate profile and AI agent to chatGPT... expand customer contact
OpenAI has released a new service that allows businesses to manage their information and communicate directly with potential customers. It is a strategy that provides corporate profiles and user analysis capabilities and expands chatGPT as…
-
Aidobi expands ‘Fier Plat’ function... Begins and Intermediate Creator Market Cho Jun
As an invasion of the creation AI, Adobe has expanded its AI platform ‘Firefly’ as an agent-based ‘AI Creative Studio’. In addition to the high threshold of the tool for professionals, it is full of strategy aimed at the early and…
-
OpenAI, quarterly sales 13th year lost 14th year...
OpenAI’s three-quarter revenue increased 2.6 times more than the previous year’s motivation, and has reached US$100 billion (about KRW 13 trillion). As the latest AI model and company’s product demand increased, growth tax recovery was…
-
Ann Tropic Launches Clone Office...Competitive to MS·Google
Andropic has created a spreadsheet and whiteboard in the Clod, co-edited and demonstrated the ability to handle tasks during video meetings. Microsoft Office (MS) or Google Workspace will be handled by the Clode itself to compete directly…
-
"One store AI avatar recommended goods". .Istsoft, Seven-Eleven and Technology Verification
Eastsoft (Representative General Manager) announced that it will test the “AI Mate” with its intelligent AI avatar technology “Ferso Interact”. AI mate is a service that helps customers interact with AI avatar on the screen and get…
-
KT provides AI mail security service to KAIST
KT(CEO Park Yunyoung) announced that it strengthens the email security system for all members such as students, researchers, and staff who provide ‘AI email security’ service to KAIST. KT AI Mail Security is a cloud security service that…
-
ChatGPT Plus for veterans and servicemembers
More than 20,000 veterans and transitioning service members have claimed a free year of ChatGPT Plus. See how they use it to prepare for civilian careers.
-
Core Weave, Sales 168% Revenue, $469 Billion...Investment Anxiety
In the last month, the coreweave urged AI-related shares to smiling market expectations, resulting in an annual view. AI infrastructure demands are still strong, but it is opaque that the investor can neglect the anxiety of the investor…
-
Participated in the development of AI factory, CJ ILILJIN
Wonframed (CEO Yoon, Kwon-dong) announced that it promotes food and beverage (F&B) autonomous manufacturing and integrating artifacts and technology development and demonstrations at the Jincheon Plant of CJ Cheil. The project will be held…
-
Nvidia employs famous law firms in New York to investigate Jenson sulfur events
Nvidia has served as a famous law firm in New York to examine the relationship between Jenson Huang CEO and the Chinese Information Agency. Judges from the former Federal Inspection will lead the investigation and take a look at how the…
- BotPsy: On-Device Personal Assistant with a Sub-Billion-Parameter Agentic Language Model
-
ScrapeBench: Evaluating Legal Compliance of AI Agents in Website Scraping
arXiv:2611.11306v2 Announce Type: replace-cross Abstract: AI agents deployed on the open web are becoming more capable of performing actions that may breach contractual provisions and violate computer security laws. In this paper, we…
-
STABLE: A Continual Learning Optimizer with Adaptive Drift Control
arXiv:2611.10655v2 Announce Type: replace-cross Abstract: Continual learning suffers from catastrophic forgetting, leading to the loss of past knowledge and unstable adaptation across task streams. Existing approaches typically address…
-
Direct Trajectory Optimization for Inference-Time Alignment of Text-to-Image Diffusion Models
arXiv:2611.11821v2 Announce Type: replace-cross Abstract: Alignment of text-to-image diffusion models with reward functions is a key challenge in controllable generation. Existing approaches typically rely on optimizing model inputs or…
-
Split and Bridge: Multimodal Generation via Diffusion Bridging
arXiv:2611.10441v2 Announce Type: replace-cross Abstract: Visual data are often expressed through multiple complementary modalities (e.g., images and segmentation masks), each capturing different aspects of the same underlying structure…
-
CulturalCrosswords: A Multilingual Benchmark of Answer Constraints and Grid Completion
arXiv:2611.10044v2 Announce Type: replace-cross Abstract: When several answers fit a riddle, exact-match scoring can penalise a valid response. CulturalCrosswords tests how answer constraints affect large language models (LLMs) solving…
-
Listen to the Features: Voice Anonymization Driven by Content Embedding Matching over Signal Reconstruction
arXiv:2607.09767v2 Announce Type: replace-cross Abstract: The paper presents a voice anonymization model focusing on preserving content rather than producing realistic speech. It relies on content embeddings extracted from a frozen…
-
Adaptive Reparametrized Time for Score-Based Diffusion Sampling
arXiv:2607.02137v4 Announce Type: replace-cross Abstract: We study timestep allocation for score-based diffusion sampling, where a learned reverse-time dynamics is discretized on a finite grid. Uniform and hand-crafted schedules are…
-
Self-Modeling Interventions can Modulate Emergent Misalignment
arXiv:2606.23700v2 Announce Type: replace-cross Abstract: Emergent misalignment (EM) has been linked to the activation of misaligned persona vectors and evil character traits, suggesting that EM operates through disruption of the model's…
-
From Architecture to Output: Testing the Structural Origins of Hallucination and the Amplifying Role of Data
arXiv:2606.07537v3 Announce Type: replace-cross Abstract: Surveys of hallucination classify errors by how an output relates to its source, and often explain them by the model's structure: attention retrieves by association, the…
-
Pruning and Distilling Mixture-of-Experts into Dense Language Models
arXiv:2605.28207v3 Announce Type: replace-cross Abstract: Mixture-of-Experts (MoE) is now the dominant architecture for frontier language models, yet it requires all expert parameters to be loaded in memory, making it less preferable for…
-
Unified Panoramic Geometry Estimation via Multi-View Foundation Models
arXiv:2605.26368v3 Announce Type: replace-cross Abstract: Geometry estimation from perspective images has greatly advanced, maturing to the point where off-the-shelf foundation models are able to reconstruct 3D scene structure not only…
-
StakeBench: Evaluating Language Understanding Grounded in Market Commitment
arXiv:2605.26074v2 Announce Type: replace-cross Abstract: Existing financial NLP benchmarks often rely on labels supplied by outside observers, measuring how language is perceived rather than what speakers have committed to in the market…
-
Diff-Instruct with Diffused Reward: Towards Principled One-step Generator RL
arXiv:2605.24001v3 Announce Type: replace-cross Abstract: Recent advances in one-step text-to-image generation have enabled real-time synthesis with remarkable efficiency and quality. Previous reinforcement learning methods for one-step…
-
Variance Reduction for Gradients from Diffusion Teachers
arXiv:2605.21489v3 Announce Type: replace-cross Abstract: Pretrained diffusion models serve as frozen teachers for pipelines such as text-to-3D, single-step distillation, and data attribution. The teacher gradients these pipelines consume…
-
Automated Root-Cause Subclassification and No-Code Fix Generation for Invalid Bug Reports
arXiv:2605.17561v3 Announce Type: replace-cross Abstract: Issues faced when using software are reported in the form of bug reports. However, many bug reports are invalid, meaning they do not require code changes, and are resolved with a…
-
AI-Mediated Communication Can Steer Collective Opinion
arXiv:2605.16245v2 Announce Type: replace-cross Abstract: Generative artificial intelligence (AI) is increasingly integrated into the online platforms where humans exchange opinions; large language models (LLMs) have been polishing users'…
-
Examining the Impact of Data and Training Methods on Emergent and Subliminal Misalignment
arXiv:2605.12798v2 Announce Type: replace-cross Abstract: Post-training a language model can induce surprising behavioral changes far beyond the intended training target. We study two particularly striking examples: emergent misalignment…
-
Minimax Rates and Spectral Distillation for Tree Ensembles
arXiv:2605.11841v3 Announce Type: replace-cross Abstract: Tree ensembles such as random forests (RFs) and gradient boosting machines (GBMs) are among the most widely used supervised learners, yet their theoretical properties remain…
-
MC$^2$: Monte Carlo Correction for Fast Elliptic PDE Solving
arXiv:2605.09288v2 Announce Type: replace-cross Abstract: Partial differential equation (PDE) solvers underpin modern computer graphics, geometry processing, and scientific computing. Classical Monte Carlo PDE solvers such as…
-
A Statistical Framework for Algorithmic Collective Action with Multiple Collectives
arXiv:2605.06749v3 Announce Type: replace-cross Abstract: As learning systems increasingly shape everyday decisions, Algorithmic Collective Action (ACA), i.e., users coordinating changes to shared data to steer model behavior, offers a…
-
AnalogRetriever: Learning Cross-Modal Representations for Analog Circuit Retrieval
arXiv:2604.23195v2 Announce Type: replace-cross Abstract: Analog circuit design relies heavily on reusing existing intellectual property (IP), yet searching across heterogeneous representations such as SPICE netlists, schematics, and…
-
Explainable Iterative Data Visualization Refinement via an LLM Agent
arXiv:2604.15319v3 Announce Type: replace-cross Abstract: Exploratory analysis of high-dimensional data relies on embedding the data into a low-dimensional space, based on which a visualization plot is produced to uncover meaningful…
-
RECAP: Local Hebbian Prototype Learning as a Self-Organizing Readout for Reservoir Dynamics
arXiv:2603.06639v2 Announce Type: replace-cross Abstract: Robust perception in brains is often attributed to high-dimensional population activity together with local plasticity mechanisms that reinforce recurring structure. In contrast…
-
Cold-Start Personalization via Bayesian Adaptive Questioning
arXiv:2602.15012v2 Announce Type: replace-cross Abstract: Cold-start personalization requires inferring preferences from minimal interaction when no user-specific historical data is available. The space of possible preferences is vast, yet…
-
BrainVista: Modeling Naturalistic Brain Dynamics as Multimodal Next-Token Prediction
arXiv:2602.04512v2 Announce Type: replace-cross Abstract: Naturalistic fMRI offers a time-resolved window into cortical dynamics during continuous multimodal experience, where future brain states are jointly governed by endogenous neural…
-
Tracing Moral Foundations in Large Language Models
arXiv:2601.05437v4 Announce Type: replace-cross Abstract: Large language models often produce human-like moral judgments, but it is unclear whether this reflects an internal conceptual structure or superficial ``moral mimicry.'' Using…
-
Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding
arXiv:2602.02977v3 Announce Type: replace-cross Abstract: Vision-language models such as CLIP often struggle to faithfully understand long, detail-rich captions, relying on dominant scene cues while overlooking fine-grained visual…
-
Pygmalion Effect in Vision: Image-to-Clay Translation for Reflective Geometry Reconstruction
arXiv:2511.21098v2 Announce Type: replace-cross Abstract: Understanding reflection remains a long-standing challenge in 3D reconstruction due to the entanglement of appearance and geometry under view-dependent reflections. In this work, we…
-
Deep Double Q-learning
arXiv:2507.00275v3 Announce Type: replace-cross Abstract: Double Q-learning is a classical control algorithm that mitigates the maximization bias of Q-learning. To do so, it explicitly trains two independent action-value functions and uses…
-
Specification-based Security Evaluation of Real-World Agents Tool-Calling under Long-Horizon Context Tasks
arXiv:2506.15253v2 Announce Type: replace-cross Abstract: As large language model (LLM) agents are increasingly deployed in long-horizon unattended scenarios, security evaluation of them before deployment becomes increasingly essential and…
-
A Computational Model of Teacher-Student Co-Adaptation
arXiv:2505.02853v2 Announce Type: replace-cross Abstract: Effective teaching involves mutual adaptation: teachers adjust instruction based on student responses, while students actively seek information to resolve uncertainty. Yet many…
-
Cross-Modality Investigation on WESAD Stress Classification
arXiv:2502.18733v2 Announce Type: replace-cross Abstract: Deep learning's growing prevalence has driven its widespread use in healthcare, where AI and sensor advancements enhance diagnosis, treatment, and monitoring. In mobile health…
-
Learning Beyond Rigid Manipulation with Hybrid Soft-Rigid Robots
arXiv:2410.07787v4 Announce Type: replace-cross Abstract: Rigid-soft robot hybrids promise to unite the precision and repeatability of rigid manipulators with the compliance and adaptability of soft structures. Yet, full-fledged rigid-soft…
-
Agentic discrete-event simulation for manufacturing system design: an open 3D simulator, a benchmark and an industrial case
arXiv:2611.10602v2 Announce Type: replace Abstract: The design of manufacturing systems relies on discrete-event simulation (DES) models that experts still build in tool-specific interfaces. AI agents can now write simulation code, but…
-
Seeing Isn't Enough: Unlocking Visual Reasoning by Bridging the Reasoning-Access Gap with TIGER
arXiv:2611.10552v2 Announce Type: replace Abstract: Multimodal Large Language Models (MLLMs) inherit strong textual reasoning capabilities from their language backbones and text pretraining, but often fail to apply the same level of…
-
Key Institutional Considerations for Designing Effective Independent Verification Organizations
arXiv:2611.10168v2 Announce Type: replace Abstract: Independent verification organizations (IVOs) have emerged as a popular, bipartisan solution for facilitating high-quality independent evaluations of AI technologies. But momentum is…
-
Scoped Semantic Invariance as a Foundation for LLM Reliability
arXiv:2610.33267v2 Announce Type: replace Abstract: LLMs often answer semantically equivalent queries inconsistently, even when the underlying facts are known. We formalize consistency as semantic invariance over equivalence classes of…
-
MoveBench: A Rubric-Based Benchmark for Evaluating Large Language Models in Sports and Health Coaching
arXiv:2610.26185v3 Announce Type: replace Abstract: Despite rapid advances in large language models (LLMs), their deployment in the sports and health domain remains high-risk due to the absence of domain-specific evaluation benchmarks…
-
Kierston & KierstonBench: Simulation of Allentown, PA and Evaluation of Language Models in Crime Investigation and Surveillance Tasks
arXiv:2611.04785v2 Announce Type: replace Abstract: Large language models are candidates for collecting and analyzing records in criminal investigations and mass surveillance. Their capability and accuracy in this work cannot currently be…
-
Probing the Origins of Reasoning Performance: Representational Quality for Mathematical Problem-Solving in RL vs. SFT Fine-Tuned Models
arXiv:2607.26119v2 Announce Type: replace Abstract: Large language models trained with reinforcement learning (RL) to reason step by step have been increasingly shown to outperform their supervised fine-tuned (SFT) counterparts on…
-
TRACE: Business Rule-Grounded Reasoning Curriculum for Knowledge-Preserving Parametric Tool Retrieval in Enterprise LLMs
arXiv:2607.22639v2 Announce Type: replace Abstract: Parametric retrieval enables LLMs to retrieve tools implicitly by assigning each API a unique virtual token and training the model to generate it via constrained beam search. Toolsense…
-
Einstein World Models
arXiv:2606.26969v2 Announce Type: replace Abstract: Does intelligence require the ability to reason about phenomena beyond direct experience? It is natural to suspect that some complex thought cannot be captured through language alone…
-
From AGI to ASI
arXiv:2606.12683v3 Announce Type: replace Abstract: Over the last decade, building human-level artificial general intelligence has moved from far-fetched speculation to being a concrete next-decade target for many of the largest AI…
-
Can AI Agents Synthesize Scientific Conclusions in Health?
arXiv:2606.11337v2 Announce Type: replace Abstract: Scientific AI agents increasingly retrieve evidence, reason across sources, and synthesize conclusions used in consequential decisions. Yet, their ability to do so in high-stakes domains…
-
Market Regime Council for Dynamic Credit Assignment in Multi-Agent LLM Decision Systems
arXiv:2605.24490v2 Announce Type: replace Abstract: Multi-agent LLM decision systems for portfolio management still lack a principled way to assign credit across specialist agents, remain vulnerable to cold-start dominance under regime…
-
Instrumental Choices: Measuring the Propensity of LLM Agents to Pursue Instrumental Behaviors
arXiv:2605.06490v3 Announce Type: replace Abstract: AI systems have become increasingly capable of dangerous behaviours in many domains. This raises the question: Do models sometimes choose to violate human instructions in order to perform…
-
TACT: Mitigating Overthinking and Overacting in Coding Agents via Activation Steering
arXiv:2605.05980v3 Announce Type: replace Abstract: When language model agents tackle complex software engineering tasks, they often degrade over long trajectories, which we define as *agent drift*. We focus on two recurring failure modes…
-
Accurate Predictions in Education with Discrete Variational Inference
arXiv:2509.23484v3 Announce Type: replace Abstract: One of the largest drivers of social inequality is unequal access to personal tutoring, with wealthier individuals able to afford it, while the majority cannot. Affordable, effective AI…
-
Identifying Macro Causal Effects in a C-DMG over ADMGs
arXiv:2504.01551v3 Announce Type: replace Abstract: Causal effect identification using causal graphs is a fundamental challenge in causal inference. While extensive research has been conducted in this area, most existing methods assume the…
-
Medical Image Segmentation: Foundation Model Adaptation versus Task-Specific Training
arXiv:2611.14519v1 Announce Type: cross Abstract: Foundation models pretrained on large natural-image collections are increasingly adapted for medical image segmentation, in the expectation that they will replace models trained for each…
-
Toward Foundation Detection Models via Example-Based Grounding in EM-DETR
arXiv:2611.14518v1 Announce Type: cross Abstract: DeepLearning (DL)-based Computer-Aided Diagnostic (CAD) systems have shown great promise in supporting clinicians, but their performance heavily depends on the availability of large…
-
One Policy for 100 Household Tasks: Our Solution for the 2026 BEHAVIOR Challenge
arXiv:2611.14432v1 Announce Type: cross Abstract: We describe our solution for the 2026 BEHAVIOR Challenge: 100 long-horizon household tasks in the OmniGibson simulator. We placed 9th on the final leaderboard, the only independent team in…
-
Two Communities, One Problem: Branch-and-Bound vs. A* for Resource-Constrained Project Scheduling
arXiv:2611.14426v1 Announce Type: cross Abstract: Branch-and-Bound (B&B) and $A^*$ are popular exact search methods originating from Operations Research and Artificial Intelligence, respectively. Although the two methods share fundamental…
-
Causal Concept Explanations for Deep Neural Models
arXiv:2611.14407v1 Announce Type: cross Abstract: Concept-based explanations in neural models are important for the development, certification, and adoption of autonomous systems. Yet, interpretability remains correlational. Without…
-
EngineAgent: Agentizing On-Device LLM Inference Engines with Inference Stack Hierarchy Awareness and Hardware Abilities
arXiv:2611.14356v1 Announce Type: cross Abstract: Rapidly evolving LLM architectures and heterogeneous mobile hardware make supporting new models in on-device inference engines a recurring and costly system challenge. Each model may…
-
AI Isn't Getting Dumber, We're Getting Pickier: Rising Standards in Human Evaluation of Large Language Models
arXiv:2611.14333v1 Announce Type: cross Abstract: Users often complain that AI assistants such as ChatGPT are getting worse. Some versions do change, but human evaluation leaves a second possibility untested: the people judging may change…
-
Ninety-Six Ways to Measure a Scribe: Fragmented Evaluation of Ambient Scribes in Healthcare
arXiv:2611.14284v1 Announce Type: cross Abstract: Ambient artificial intelligence (AI) scribes have moved rapidly from experimental pilots to enterprise-wide deployment across health systems, promising reduced documentation burden and…
-
Lilo: A Multi-Agent Conversational Sandbox for Children's Digital Communication
arXiv:2611.14260v1 Announce Type: cross Abstract: Children's peer conflicts increasingly play out in group chats, where tone and timing are stripped away, and a hurtful message is easier to send than to say aloud. We present Lilo, a…
-
Dyship: Leveraging Acoustic Features For Speaker Relationship Classification From Dyadic Conversations
arXiv:2611.14241v1 Announce Type: cross Abstract: We introduce DyShip, a framework for classifying speaker familiarity and kinship from dyadic conversations in the Seamless Interaction corpus. We compare five speech representations…
-
Verifiable Agentic Traffic Class: A Policed Quality-of-Service Mechanism for Agentic AI Traffic, Evaluated from Simulation to Real-Kernel Adversarial Testing
arXiv:2611.14211v1 Announce Type: cross Abstract: Software agents built on large language models generate traffic unlike human web traffic, yet routers have no class for it and cannot verify a sender's claim to be an agent. This paper…
-
Representational Atoms in Neural Networks
arXiv:2611.14168v1 Announce Type: cross Abstract: Understanding the internal representations of neural networks and characterizing their relationship with human conceptual structure are two critical questions for elucidating model…
-
Visual Similarity in Adult Word Extension and Artificial Word Learning
arXiv:2611.14116v1 Announce Type: cross Abstract: When two objects share a name, how much does their visual similarity explain, and what do the teaching examples add? First, we reanalyzed trial-level data from a large adult word-extension…
-
Looking One Step Ahead: An Anticipation-guided Agent for Object State Readout in Videos
arXiv:2611.14113v1 Announce Type: cross Abstract: Understanding the progress of an activity requires tracking how the objects involved change over time. Object state readout provides this information by describing a target object's states…
-
Model-Free Runtime Measurement of Task-Relevant Information
arXiv:2611.14111v1 Announce Type: cross Abstract: Autonomous systems increasingly monitor physical kinematics, environmental states and task completion outcomes. While existing work provides methods to plan information seeking actions and…
-
ProofCoder: Continuous Reasoning for Automated Software Verification
arXiv:2611.14106v1 Announce Type: cross Abstract: Formal verification using proof assistants provides strong correctness guarantees for software, but constructing proofs remains labor-intensive and requires significant expertise. Recent…
-
From Bias to Positionality: A critical grammar for Large Language Models
arXiv:2611.14079v1 Announce Type: cross Abstract: Bias has become a default lens for understanding what machine learning systems do and how we should relate to them. Bias is at root a deviation from a standard; in decision making systems…
-
ZoomGain: Mining Global-to-Zoom Perception Gaps for Fine-Grained Visual Supervision
arXiv:2611.14076v1 Announce Type: cross Abstract: Fine-grained visual perception requires Vision-Language Models (VLMs) to locate and interpret subtle evidence within complex scenes. Constructing supervision for this ability is…
-
Obscured Wildfire Flame Detection by Spatio-temporal Analysis of Smoke Patterns Using Frame-wise Transformers
arXiv:2611.14074v1 Announce Type: cross Abstract: In this paper, we propose a new paradigm in wildfire management for detecting both visible and obscured fire using Deep Learning (DL). The essence of our method is the joint spatial and…
-
HandsOnAid: Building and Evaluating an Offline First-Aid Assistant for Emergency Response Scenarios
arXiv:2611.14059v1 Announce Type: cross Abstract: Emergencies can occur where there is no phone signal and no ambulance nearby. A small language model on a phone can be helpful in such situations, but its advice needs to be safe and…
-
Can KV Cache Be Reused Across Model Updates? A DeepSeek-V4-Flash Case Study
arXiv:2611.14051v1 Announce Type: cross Abstract: Reusing key--value (KV) caches across language model updates could avoid repeated computation of shared context, but changes in model weights raise a fundamental question about what these…
-
MISO-BR: An Open Corpus for Misogyny Detection in Brazilian Portuguese
arXiv:2611.14036v1 Announce Type: cross Abstract: Misogyny is among the most prevalent forms of online hate speech directed at Brazilian users, yet open and auditable resources dedicated to it in Brazilian Portuguese remain scarce. We…
-
Few-Step Streaming Video Generation via Condition Shifting
arXiv:2611.14033v1 Announce Type: cross Abstract: Few-step sampling reduces the cost of streaming video generation. Distribution Matching Distillation (DMD) supervises few-step generators with a separate fake-score network, which adds…
-
Program-Aided Patchable Skill Evolution
arXiv:2611.14019v1 Announce Type: cross Abstract: Agent skills have emerged as an effective means of enabling LLM agents to solve complex, interactive tasks by encoding reusable knowledge without modifying model parameters. Recent…
-
Margin-Guided Token Credit for Reinforcement Learning with Verifiable Rewards
arXiv:2611.14015v1 Announce Type: cross Abstract: In reinforcement learning with verifiable rewards, a language model is rewarded for a correct final answer, and GRPO passes that reward to every token of the solution in equal measure. We…
-
Coding Agent Memory Post-Training II: Unlocking the Memory Potential of Pre-trained File Operations for Long-Horizon Tasks via On-Policy Distillation
arXiv:2611.14009v1 Announce Type: cross Abstract: Language-model agents increasingly tackle long-horizon tasks whose interaction histories exceed the model's active context. They must therefore decide what to write down and what to read…
-
TransGHOST: Real-Time, Synthesis-Free Reconstruction of Instrument-Occluded Regions in Ophthalmic Microsurgery
arXiv:2611.13971v1 Announce Type: cross Abstract: Intraoperative scene perception in ophthalmic microsurgery is limited by instrument occlusion: when a tool enters the high-magnification field, it conceals tissue from both the surgeon and…
-
Controlled Knowledge Engineering for Taxonomy Induction with Tool-Grounded LLM Agents and Runtime-Adaptive Policies
arXiv:2611.13966v1 Announce Type: cross Abstract: Taxonomies are structured knowledge representations that support semantic search, recommendation, product catalog organization, and retrieval-augmented generation. Their automated…
-
From Attack Surface to Defense Architecture: Compositional Security for Agentic Coding Harnesses
arXiv:2611.13960v1 Announce Type: cross Abstract: AI coding harnesses couple language models to shells, filesystems, and network sockets, turning the harness into an automated Living-off-the-Land (LotL) platform: once an adversary injects…
-
Chem World: A Large-Scale Benchmark and Physics-Aware Framework for Trustworthy Molecular Property Prediction
arXiv:2611.13945v1 Announce Type: cross Abstract: Molecular property prediction is fundamental to computational chemistry and drug formulation research, yet existing benchmarks remain fragmented across datasets, properties, and evaluation…
-
Tailoring Optimization Approaches with Domain Knowledge in Industry Process for Meta-heuristic Development
arXiv:2611.13940v1 Announce Type: cross Abstract: The paper presents an Industry Process for Meta-heuristic Development (IPMD) schema, a systematic methodology for applying meta- and hyper-heuristic applications to optimization problems…
-
The Case for AI-Enabled Gate Monitoring as Farm Security in the UK: A PRISMA-Guided Review of the Evidence on Rural Crime, Livestock Predation and Farm Access
arXiv:2611.13933v1 Announce Type: cross Abstract: Farm security in the UK rests on thinner evidence than its cost suggests. We ran a PRISMA guided review: twelve searches of Web of Science and Scopus on 2nd -4th November 2026 returned…
-
Domain-Invariant and Language-Model-Assisted Acoustic Side-Channel Attacks on Keyboards
arXiv:2611.13922v1 Announce Type: cross Abstract: Acoustic side-channel attacks (ASCA) recover typed text from the sound of keystrokes, using only a commodity microphone and no access to the victim's device. Prior work establishes that…
-
CAppleNet: A Coordinate Attention-Based YOLO11 for Real-Time Apple Stem Cavity and Calyx Detection
arXiv:2611.13920v1 Announce Type: cross Abstract: Automated fruit grading, sorting, and orientation systems require real-time, accurate detection of fruit-level morphological landmarks rather than the fruit boundary alone. This paper…
-
MicroBT: Centroid-Guided Synergistic Learning for Multi-Modal Micro Brain Tumor Segmentation and Counting
arXiv:2611.13912v1 Announce Type: cross Abstract: Multi-modal brain tumor segmentation is essential for clinical diagnosis, radiotherapy planning, and longitudinal monitoring. While recent studies have achieved remarkable overall…
-
Transparency Disclosures for Chain-of-Thought Monitorability
arXiv:2611.13873v1 Announce Type: cross Abstract: Chain-of-thought (CoT) monitoring is one of the most valuable tools for monitoring advanced AI systems, but it is under threat. We argue that frontier AI developers should report specific…
-
Mechanistic Interpretability of Deep Neural Networks: Methods, Frameworks, and Perspectives
arXiv:2611.13868v1 Announce Type: cross Abstract: Deep neural networks underpin fundamental tasks in a wide range of signal domains, such as image, audio, and language processing. However, their internal mechanisms are notably difficult to…
-
PolyRoute: A Twelve-Language and Code-Mixed Benchmark for Typed-Decision Routing, with Baselines and a Leaderboard
arXiv:2611.13843v1 Announce Type: cross Abstract: Typed-decision encoders answer a question about a text in one forward pass, but their behaviour outside English is rarely measured on a fixed yardstick. We release PolyRoute: 322…
-
MusicFlux: Flexible Conditioning for Text-Guided Music Generation and Editing
arXiv:2611.13834v1 Announce Type: cross Abstract: Music editing transforms existing recordings while preserving their original attributes. Recent approaches adapt pretrained text-to-music models for tasks like inpainting, event insertion…
-
CASCADE: Constraint-Aware SCAffolded DEcomposition for Logical Anomaly Detection
arXiv:2611.13820v1 Announce Type: cross Abstract: Logical Anomaly Detection (LAD) requires verifying whether all category-specific constraints, such as object counts, pairing relations, and measurement ranges, are satisfied by a query…
-
Swanada: Split-Encoder Capability-Routed Conformer CTC and Sub-30 MB On-Device Speech Recognition Across Twelve Scripts
arXiv:2611.13814v1 Announce Type: cross Abstract: Client-side automatic speech recognition (ASR) across Indic and Classical scripts eliminates cloud transcription costs, network tail latency, and server-side audio retention, yet neural…
-
Bridging Graph and Pairwise Learning for Gene Regulatory Network Inference from scRNA-seq Data
arXiv:2611.13801v1 Announce Type: cross Abstract: Motivation: Gene regulatory networks (GRNs) capture the underlying interactions through which transcription factors (TFs) regulate genes. Based on gene expression data, existing GRN…
-
Softness is something you also need! Why compliance helps control: information metrics of a passive walker with compliant ankle
arXiv:2611.13779v1 Announce Type: cross Abstract: A great gap in adaptivity, flexibility, and dexterity still exists between intelligent robots, whether they are made of rigid or (partially) compliant components, and their naturally…
-
DisasterAgent: An Evidence-Driven Agent for Verifiable Disaster Report Generation From Remote Sensing Imagery
arXiv:2611.13776v1 Announce Type: cross Abstract: Remote sensing disaster assessment is moving from simple damage classification toward comprehensive report-level analysis. In this setting, responders need not only damage descriptions but…
-
Mixed Signals: Weight-Space Programming for EEG and ECG
arXiv:2611.13758v1 Announce Type: cross Abstract: Weight-space programming allows language models to read a trained neural network's weights and edit them. This opens the door to ``training'' by reasoning rather than through gradient…
-
On Private Knowledge Retrieval and Private Inference for secure multi-agent systems
arXiv:2611.13696v1 Announce Type: cross Abstract: Multi-agent systems provide a convenient framework for the massive deployment of agents based on Large Language Models (LLMs). It provides a sound framework where agents can interact with…
-
AURA: Auditory Spatial Intelligence via a Unified Representation from a General-Audio Encoder through Complete Alignment
arXiv:2611.13687v1 Announce Type: cross Abstract: Answering spatial questions about multi-source scenes requires a model both to identify each sound and to locate it. Existing spatial audio-language models obtain spatial information from a…
-
Ego Guidance: Enhancing Robot Policy Learning with Action-Free Human Egocentric Videos
arXiv:2611.13685v1 Announce Type: cross Abstract: In this paper, we investigate how human egocentric videos can benefit robot policy learning without action supervision. Building on a pretrained VLM, we propose Shared Prediction on Action…
-
MOAI: Module-Optimizing Architecture for non-interactive secure transformer Inference
arXiv:2611.13641v1 Announce Type: cross Abstract: Privacy concerns have been raised in Large Language Models (LLM) inference when models are deployed in Cloud Service Providers (CSP). Homomorphic encryption (HE) offers a promising solution…
-
Synapse-Control: Making Agent Experience Actionable with Lightweight Runtime Control
arXiv:2611.13615v1 Announce Type: cross Abstract: Large language model agents reuse execution experience to improve reliability on complex tool-based tasks. Systems such as Sentry call large models to generate failure diagnoses before…
-
PhyRubric: Rubric-Grounded Diagnostic Evaluation of Physical Commonsense in Text-to-Video Generation
arXiv:2611.13576v1 Announce Type: cross Abstract: As text-to-video (T2V) models are increasingly positioned as world simulators, it is essential to evaluate whether their videos conform to physical commonsense. However, existing benchmarks…
-
Market-Conditioned JEPA: Learning from Unpredictable Market Shocks
arXiv:2611.13572v1 Announce Type: cross Abstract: Much of the variation in financial time series is driven by market-wide shocks that cannot be predicted from an asset's own history. Per-asset self-supervised learning treats this variation…
-
A Concept-Based Interpretable Framework for Clinical Text-based Prediction via Token-Concept Attention without Concept Supervision
arXiv:2611.13564v1 Announce Type: cross Abstract: Objective: Concept-based interpretation has emerged as a promising approach for explaining blackbox models using human-understandable, high-level concepts. However, most existing methods…
-
Massive Activations Confound Sparse Autoencoders
arXiv:2611.13548v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) are commonly evaluated by variance explained and ablation damage. We show how massive activations undermine both evaluations, and propose a correction. Massive…
-
Developing an Operational Model for an Internet of Data Based Trusted Data Space for Vocational Education
arXiv:2611.13520v1 Announce Type: cross Abstract: Data circulation in vocational education has long been constrained by unclear ownership, a lack of trust, and insufficient incentives. The Internet of Data provides a new underlying…
-
Sharp Sample Complexity of KL-Regularized Reinforcement Learning
arXiv:2611.13480v1 Announce Type: cross Abstract: KL-regularized reinforcement learning, the objective behind maximum-entropy RL and RL from human feedback, is variational inference in the graphical model of "control as inference": the…
-
ForeSpark: Joint Draft and Verification Budget Scheduling for Parallel Speculative Decoding
arXiv:2611.13442v1 Announce Type: cross Abstract: Parallel drafters accelerate LLM decoding by proposing a whole block of draft tokens in a single forward pass, making them attractive for LLM serving. Recent systems optimize goodput (i.e…
-
You Don't Need Feature Learning to Do In-Context Learning
arXiv:2611.13431v1 Announce Type: cross Abstract: How should we understand memorization vs. generalization in in-context learning (ICL) tasks, and when does one happen vs. the other? To investigate, we study how rotation-invariant kernel…
-
Free Lunch Uncertainty: Sequence-Level R\'enyi Entropy from Particle Decoding for Adaptive LLM Reasoning
arXiv:2611.13414v1 Announce Type: cross Abstract: Quantifying uncertainty over an entire LLM completion is critical for selective prediction and adaptive compute. Zero-cost signals such as mean token entropy and log-probability are local…
-
Placing LLM Lookup Tables by Per-Layer Prefetch Deadlines
arXiv:2611.13413v1 Announce Type: cross Abstract: DeepSeek-V4.1's 203 GB of Engram n-gram tables sit in host DRAM today, taking 190 of the 223 GiB a 4-GPU serving process uses. Row indices depend only on input tokens, so reads can start…
-
Stinging the Swarm: Extracting Attacker Intelligence from a Twin-Controller OT Honeypot
arXiv:2611.13406v1 Announce Type: cross Abstract: Autonomous language-model agents can drive industrial controllers over fieldbus protocols, probing a plant at machine speed and in coordinated swarms. We present Talos, an inline gate that…
-
The Dose Makes the Poison: Exposure-Aware Evaluation of Generative Drug Design
arXiv:2611.13398v1 Announce Type: cross Abstract: Generative models are increasingly used to propose molecules for drug discovery. They can produce far more candidates than can be experimentally tested, making computational selection…
-
The AI Era's Lines of Code? A Case Study on What an Engineer's Token Spend Measures
arXiv:2611.13396v1 Announce Type: cross Abstract: Organizations use engineers' token spend to monitor AI adoption, set budgets, and measure productivity, but the metric's relationship with engineering output is unclear. We link five weeks…
-
Enforcing Axioms for AI Alignment under Loss-Based Rules
arXiv:2611.13369v1 Announce Type: cross Abstract: Many alignment methods, most notably reinforcement learning from human feedback (RLHF), train reward models by minimizing a loss on pairwise comparisons of model responses. Inspired by…