Results for “model”
Multi-HMM Embeddings for Dynamic Malware Classification Using API Call Sequences
Classifying malware into families and categories remains difficult because modern threats are complex and polymorphic. This study couples Hidden Markov Model (HMM)-based representations of Application Programming Interface (API) call sequences with supervised learning. A class-specific HMM is trained for each malware class, and four descriptors (state occupancy, transition, observation compatibility, and likelihood), extracted from the posterior quantities of every HMM, are…
Priced at disclosure, revealed at horizon: cybersecurity incidents and deferred market risk
Purpose This study examines whether cybersecurity incident disclosures are associated with short-window market valuation and subsequent risk measures of large listed Indian firms. Design/methodology/approach A market-model event study evaluates 29 retained incidents involving NIFTY 50 constituents. A quarterly panel of 47 firms and 3,002 firm-quarter observations is then used to estimate post-incident changes in cost of equity and realised volatility through two-way…
Optimal party strategies on niche issues: a computational approach to multiparty competition for position and salience
Abstract How should parties respond to competition from a new challenger on a niche issue? Meguid argues that accommodation is effective, but recent evidence shows it may cost votes. We identify vote-maximizing strategies of established parties facing an extreme challenger using an extended Downsian model with position and salience. Based on empirical data from Germany, our results show that accommodation works only for low-valence parties, and only while the challenger is…
Practical Deep Learning-Based Block Demapping for mmWave MIMO Systems
Deep learning is expected to play a central role in 6G systems, by enabling data-driven physical layer designs that can adapt to practical wireless impairments. This is particularly important for millimetre-wave (mmWave) multiple-input multiple-output (MIMO) systems, which are strongly affected by residual channel distortions, interference, and hardware non-idealities that are difficult to capture with analytical models. In this paper, we propose a deep learning-based soft…
Same Error, Different Recovery: How Delivery Changes LLM Agent Behavior
A tool-using language model must decide whether to retry a failed call, correct it, or stop and tell the user. The error text alone may not be enough: the same feedback can lead to different actions depending on where it appears in the conversation. We isolate this delivery problem by varying placement, guidance, retained history and turn structure. Two studies with plans fixed before data collection, across 13 open models from 3B to 70B parameters, compare the next action…
Continuous modeling of manipulative participation for stance analysis: Online discourse on the 2024 Street Animal Law in Türkiye
Studies of online political discourse commonly infer public opinion through stance analysis while implicitly assuming that observed user participation is authentic. Meanwhile, studies focusing on detecting manipulative participation are typically modeling it as a binary phenomenon, overlooking the varying degrees to which users may engage in coordinated or inauthentic behavior. This study proposes a continuous modeling approach for manipulative participation and examines its…
Treating the Model as an Untrusted Component: Deterministic Policy Enforcement, Human Approval and Taint Tracking Against Prompt Injection in Tool-Using Agents, with Validation on Open-Weights Models
Tool-using LLM agents act on instructions embedded in the data they read, so prompt injection is a structural rather than a prompt-wording problem. We compare in-prompt defences (hardened prompts, spotlighting, LLM detectors) with deterministic enforcement outside the model (four reference monitors of increasing strength), human approval and taint tracking, judged by one effect-based oracle independent of every defence in an executable environment of 13 tasks and 12 tools…
Safeguarding Agentic AI in Radiology Workflows: A Multimodel Evaluation of Rule-based and Large Language Model-based Verification
Purpose: To estimate how often tool-using large language model (LLM) agents record clinically relevant false or unsupported information in simulated radiology documentation and how much safeguards applied before each action reduce these errors. Materials and Methods: In this simulation study (October 2026), 48 synthetic, radiologist-reviewed base scenarios covered finding communication, tumor board preparation and follow-up management. Forty-three also had a…
On the Relative Security of Prompt Injection: Mechanised Definitions, Defences and Verdicts for LLM Agents
An LLM agent that reads untrusted content, such as an email or a web page, may follow instructions hidden in it. Noninterference is too strong for such agents, since a useful agent lets untrusted content influence what it does and may disclose the user's data by design, and declassification needs a release policy written for each task. We instead state security relative to a reference agent, which runs the same plan under the same runtime monitor with a model that never…

Moving from AI adoption to an AI operating model: How Docusign engineering went beyond tool rollout
Docusign rolled out AI coding tools across its engineering teams, expecting them to help developers write code faster and do
Learning through Capacity Awareness to Minimize Hospital Readmissions
Predictive algorithms are increasingly used to guide hospital discharge decisions, yet these models can exhibit censoring bias for vulnerable patients due to insufficient historical data or past discharge decisions. When biased readmission risk predictors are used to inform discharges, the training data generated by discharge decisions may perpetuate similar biases when predictors are re-trained, leading to a vicious cycle. We extend a queueing-based Markov decision process…
CU-3D: A Scalable Legality-Aware RTL-to-GDS Flow for 3D-IC Physical Design
Face-to-face (F2F) hybrid bonds shorten interdie connections, but data, clock, and power-grid (PG) connections must be realized together. CU-3D is a RTL-to-GDS physical design flow based on commercial EDA tools. It derives role-compatible hybrid-bond (HB) sites from a 17-metal ASAP7-based technology model, complete HB footprints, and constructed PG geometry. To coordinate partitioning with placement flexibility, Scaled refines an inherited 2D placement, whereas Fresh places…

Apple’s M5 MacBook Air is down to its lowest price since July
The latest 13-inch MacBook Air is currently available for $1,099.99, marking the second time we’ve seen this $200 discount following the company’s hefty price increases over the summer. Although the base model — which features Apple’s M5…
When Coding Agents Get Arms: Safety Risks Across the Digital-Physical Boundary
Advances in spatial understanding have enabled frontier models to power coding agents that control a real robot arm zero-shot and end-to-end, without task-specific demonstrations or retraining. This brings two new safety challenges. First, the extensive knowledge of frontier models lets these agents perform tasks that earlier VLA and WAM paradigms could not, opening the door to new forms of capability misuse. Second, coding agents can access digital resources such as web…
Small Data Generative AI for Culturally Specific Artistic Creation: A Case Study on the Sonic Space of sarir
Large-scale generative AI (GenAI) models have become increasingly integrated into contemporary creative practice, yet they often reproduce cultural biases and fail to meaningfully engage with non-Western artistic traditions. In response, this paper explores a small-data approach that foregrounds artist-curated and culturally informed dataset creation and model training. Through a case study on sarir (the distinctive sound produced by the reed pen in Persian calligraphy), we…
DIY Customized Speech Recognition for a Person with Speech Disabilities
General-purpose state-of-the-art automatic speech recognition (ASR) systems fail systematically for speakers with moderate to severe speech disabilities. This paper describes the design, implementation, and evaluation of a customized ASR system built for the first author, a researcher with cerebral palsy and dysarthria. We fine-tuned OpenAI's Whisper medium and large-v3 models using Low-Rank Adaptation (LoRA) on audio samples recorded from the first author's own speech on…

Scientists simulate an Earth without life to help hunt for aliens and inhabited worlds
By modeling how a planet without life develops its atmosphere, astronomers hope to be able to better distinguish rocky planets without life from those that are inhabited.
Google AI Releases Elephant-Goldfish Model Plugin for Spec-First Coding
Google's Elephant-Goldfish Model plugin makes coding agents draft specs, then runs 5 zero-context review stages before any code.
Which Routing Does Each MoE Model Use? DeepSeek, Qwen, Mixtral, Llama 4, gpt-oss Compared
A config-verified comparison of how 10 open MoE models route tokens: total and active parameters, expert counts, top-k, shared experts, gating functions and load-balancing methods.

World Models Will Soon Be Inescapable. Here’s What You Need to Know
Moving AI from text to 3D space, world models simulate real-world physics to teach robots and autonomous cars how to navigate reality.

The Apple MacBook Pro M5 is on Sale, and a Top MacBook to Buy Right Now
Apple announced an updated 14-inch MacBook Pro with the latest M6 chip. But with the M5 model on sale, I’m not sure the M6 is worth the upgrade.

AI coding agents still need human cleanup on nearly three out of four tasks in a real-world test
Alibaba's new coding-agent benchmark, Co-Design Bench, tests AI agents on real user requests that already tripped them up in practice. Even GPT-6 Astra, the top performer, fully solves only 26.4 percent of tasks. On older benchmarks, the same models score above 99 percent. The article AI coding agents still need human cleanup on nearly three out of four tasks in a real-world test appeared first on The Decoder.
ICLayout-Bench: Benchmarking LLM Agents for Custom Circuit Layout Design
Large language model (LLM) agents are being explored for circuit design, but their ability to produce high-quality custom layouts has not been extensively evaluated. ICLayout-Bench is a benchmark in which agents use electronic design automation (EDA) tools to construct and optimize GDSII layouts from transistor-level circuit netlists. These layouts are evaluated using physical verification, parasitic extraction, and post-layout simulation. Layouts that pass the physical and…
Do Benign-Trained Hidden-State Probes Detect Tool Errors Under Adversarial Conditions?
Although hidden-state probes efficiently monitor model operation, it remains unclear whether probes trained on benign tool use remain useful for detecting incorrect actions under adversarial conditions. In this research, we assess this problem from the perspective of operational tool-action correct- ness. we use four AgentDojo suites, four conditions, and three models: Qwen2.5-7B-Instruct, Qwen2.5-14B-Instruct, and GLM- 4.6V-Flash (9B). We perform a five-fold, task-grouped…
MoE Routing Strategies Explained: Top-K, Expert-Choice, Sigmoid Gating and Dynamic Routing (2026 Guide)
MoE routing strategies decide which experts process each token. This guide explains 7 methods, from Mixtral's top-2 softmax to DeepSeek-V4's hash routing, with the models that use each one.
Efficient VLA Inference via Retrieval-Based Speculation and Hardware Acceleration
Vision-language-action (VLA) models enable robots to understand and execute complex instructions, but their autoregressive (AR) action generation introduces significant latency on resource-constrained robotic platforms. Speculative decoding can accelerate AR-based inference by parallelizing action generation and verification; however, existing approaches rely on costly draft models, which are inefficient on edge platforms. In this paper, we propose Zora, a hardware-software…
When Is High-Bandwidth Flash Cost-Effective for LLM Inference?
Can high-bandwidth flash (HBF) make LLM inference cheaper than adding GPUs? By storing weights closer to the GPU than SSD, HBF may reduce the number of GPUs needed for model capacity. However, this integration is only justified if it meets service targets and its cost is recovered. We simulate this integration using public component data across nine dense and mixture-of-experts (MoE) models, five workloads, two GPU platforms, and 1--100 GPUs. At fixed demand, HBF reduces the…
Biosecurity in Laboratory Animal Facilities Supporting Agricultural Research: Concepts and Practical Applications
Biosecurity in laboratory animal facilities supporting agricultural research is often viewed as a means of preventing the introduction, transmission, or release of infectious agents. These functions remain essential in facilities housing livestock, poultry, aquaculture species, and laboratory animal models used to investigate agricultural priorities. Such research supports the development and evaluation of veterinary medicines, diagnostic methods, feed and nutrition…
Monocular Metric-Depth Adaptation for Agricultural Perception: A CitrusFarm Case Study with Embedded Inference
Adapting monocular metric depth to agricultural imagery requires balancing target-domain accuracy against the scope of model updates and embedded execution cost. We evaluate SARE-Depth on CitrusFarm using a frozen teacher-guided Depth Anything V2 Small base. Depth-Preserving Residual Head Refinement (DP-RHR) learns image-level log-scale and gated local corrections, followed by fixed depth-conditioned Range-Expert (RE) refinement. Before adaptation, fixed-checkpoint…