small language model

Improving Medical Calculation of LLMs with Embedded Coding

Large Language Models (LLMs) perform well on medical examinations and question-answering benchmarks, but remain unreliable on medical calculation tasks that require exact numerical outputs. These calculations support high-stakes decisions such as medication dosing, organ-function assessment, and prognostic scoring, for...

Translating Post-Quantum Cryptography Roadmaps into an Actionable SME Migration Framework

The results show that PQC migration is not simply an algorithm-replacement exercise; it is an organisational transformation process requiring governance, cryptographic visibility, vendor coordination, phased implementation, and continuous monitoring.

Human-in-the-loop formative assessment with smaller language models: An equivalence testing study in programming education

Role of artificial intelligence in the diagnosis of amyotrophic lateral sclerosis via biomarkers, neuroimaging, and wearable health technologies: a narrative review

Artificial intelligence-enabled approaches show promise for earlier and more objective ALS detection, but the current literature is dominated by small, retrospective, single-site studies.

Pwnagent: a knowledge-guided multi-agent system for automatic exploit generation

PwnAgent, an LLM-driven multi-agent framework for end-to-end exploit generation that combines offensive domain knowledge with active runtime introspection, is presented and paired results indicate that structured knowledge guidance, execution-grounded measurement, and feedback repair improve LLM-based exploit generatio...

MIGServe: Layout-Aware Multi-Instance GPU Management for Efficient LLM Serving

MIGServe treats the physical layout of MIG instances as a first-class scheduling dimension through three techniques: buddy-aware partition placement, which preserves large contiguous free blocks by allocating next to existing occupied buddies; proactive pair-matching migration, which consolidates fragmented half-full b...

Edge-Cloud Collaborative RAG with Multi-State Semantic Caching

A Multi-State Semantic Cache (MSSC) framework that dynamically maps knowledge chunks into three mutually exclusive states: zero cache, index cache, and full cache is proposed that reduces average service latency by over 40% in dynamic scenarios and suppresses backhaul reconfiguration traffic by up to 38% compared to co...

CARE-MoE: Correlation-Aware Expert Placement and Semantic Equivalence Routing for MoE LLM Inference on Edge Devices

CARE-MoE is proposed, an efficient MoE LLM inference framework comprising two core components that balances expert placement by jointly modeling co-activation correlation and hot–cold drift, preventing overload from correlated experts and enabling low-cost adaptive rebalancing.

Region-Level Black-Box Defense Against Stealthy Embedding-Space Backdoors in CLIP

CLIPGuard is proposed, a lightweight and fully black-box defense specifically designed to mitigate embedding-space backdoors in CLIP encoders, consistently outperforming existing black-box defenses, including CleanCLIP and CleanerCLIP.

Configuration, Not Conscience: A Large-Scale Empirical Study of LLM System Prompts

Analysis of leaked system prompts from 62 vendors across four community collections supports treating leaked prompts as operational specifications, closer to configuration files than value statements, and treats reuse and prompt rot as engineering and supply-chain concerns.

Cache-Fused Kinematic Rails: Ex-Ante Silicon Alignment via In-Vivo KV-Cache Metric Grafting

Abstract of Cache-Fused Kinematic Rails: Ex-Ante Silicon Alignment via In-Vivo KV-Cache Metric Grafting Current Large Language Model (LLM) alignment relies primarily on post-hoc, output-level correction mechanisms such as external API meta-governors, Reinforcement Learning from Human Feedback (RLHF), or post-generation...

Reproducible Generative Constraints in Voynichese

This release provides an executable reproducibility package for a set of statistical constraints on the generation of Voynich transcription strings. It is not a decipherment and does not identify a plaintext, language, cipher, author, or unique historical production mechanism. The aim is narrower: to define measurement...