research

LLM Security

STAR Method Detects Hidden Backdoors in LLM Reasoning Chains

New research introduces State-Transition Amplification Ratio (STAR) to identify inference-time backdoor attacks in large language models by analyzing anomalous reasoning patterns.

LLM Infrastructure

AIConfigurator Speeds Up LLM Serving Optimization Dramatically

New research introduces AIConfigurator, a system that dramatically accelerates configuration optimization for multi-framework LLM serving, enabling faster deployment of AI inference infrastructure.

LLM Security

New Research Exposes How LLMs Fall for Fake Evidence

Researchers reveal how large language models can be manipulated with fabricated evidence, raising critical questions about AI reliability and the spread of misinformation through synthetic content.

LLM Quantization

FLRQ: Faster LLM Quantization via Low-Rank Matrix Sketching

New quantization method FLRQ achieves up to 2.5x faster compression of large language models while maintaining accuracy through flexible low-rank matrix approximation techniques.

LLM

Proactive Memory Extraction Advances LLM Agent Capabilities

New research proposes proactive memory extraction for LLM agents, moving beyond static summarization to enable more dynamic knowledge retention and recall in autonomous AI systems.

LLM unlearning

Dual-Granularity Data Synthesis Advances LLM Unlearning Methods

New research introduces domain-to-instance framework for generating synthetic data to help large language models selectively forget harmful knowledge while preserving useful capabilities.

AI Safety

GuardEval: New Benchmark Tests LLM Content Moderators

Researchers introduce GuardEval, a comprehensive benchmark evaluating LLM moderators across safety, fairness, and robustness dimensions—critical metrics for AI content authentication systems.

Explainable AI

xDNN(ASP): Logic-Based Explainability for Neural Networks

New research combines deep neural networks with Answer Set Programming to generate human-readable explanations for AI decisions, advancing interpretability crucial for detection systems.

LLM

Multi-Agent System Automates LLM Prompt Optimization

New research introduces an evaluation-driven multi-agent workflow that automatically optimizes prompt instructions for improved LLM instruction following performance.

AI Certification

Embodied AI Certification Framework Proposes Trust Metrics

New research proposes maturity-based certification for embodied AI systems, introducing quantifiable trustworthiness metrics that could reshape how we evaluate AI reliability and authenticity.

AI Security

Privacy Attacks Target Graph Diffusion Models: New Security Risks

New research reveals three classes of inference attacks against graph generative diffusion models, exposing membership inference, property inference, and data reconstruction vulnerabilities in AI generation systems.

LLM Optimization

LLM Agent Automates Hardware-Aware Model Quantization

New research introduces an LLM-based agent that automatically selects optimal quantization strategies for deploying large language models across diverse hardware platforms.