Machine Learning

AI Safety

New Framework Certifies AI Agent Reliability Without Model Access

Researchers propose combining self-consistency sampling with conformal calibration to certify AI agent reliability without requiring access to internal model weights or architecture details.

neural networks

SymTorch: Extracting Symbolic Math from Neural Networks

New framework converts opaque neural network decisions into interpretable mathematical expressions, enabling better model verification and understanding of AI behavior.

content moderation

ML Sampling + LLM Labeling: A New Framework for Content Moderatio

New research proposes combining ML-assisted sampling with LLM labeling to measure policy-violating content at scale, offering a methodological breakthrough for detecting synthetic media and deepfakes.

AI Agents

5 Essential Metrics for Evaluating AI Agent Performance

Moving beyond simple accuracy, these five metrics—task success rate, tool usage accuracy, context coherence, response latency, and safety compliance—reveal what truly matters when assessing AI agents.

LLM Interpretability

ADAPT: Hybrid Prompt Optimization Advances LLM Interpretability

New research introduces ADAPT, a hybrid optimization technique that combines discrete and continuous methods to visualize and understand internal features of large language models.

LLM fine-tuning

Influence-Preserving Proxies Accelerate LLM Fine-Tuning Data Sele

New research introduces proxy methods that preserve gradient influence signals while dramatically reducing computational costs for selecting optimal training data in large language model fine-tuning.

LLM

Variability Modeling Meets LLMs: Tuning Inference Parameters

New research applies software product line variability modeling to systematically optimize LLM inference hyperparameters like temperature and sampling strategies.

AI Security

IARPA TrojAI Program: Detecting Hidden Backdoors in AI Models

IARPA's TrojAI program releases final report on detecting trojan attacks in AI systems, covering image classifiers, NLP models, and reinforcement learning with implications for synthetic media security.

AI Agents

How to Build Self-Improving AI Agents Using Langfuse

Learn how to create AI support agents that continuously improve through feedback loops using Langfuse observability. A technical guide to building autonomous systems that learn from interactions.

LLM research

Benchmarking Uncertainty Metrics in LLM-Based Assessment Systems

New research introduces a comprehensive benchmark for evaluating how well LLMs can quantify their own uncertainty when grading, with implications for AI reliability and trustworthy automated systems.

multimodal AI

5 Multimodal AI Architectures Powering Video and Image AI

From early fusion to cross-modal attention, understanding the five core architectures behind AI systems that can see, read, and understand simultaneously—the foundation of modern synthetic media.

LLM fine-tuning

Zero-Order Optimization Enables Memory-Efficient LLM Fine-Tuning

New research introduces learnable direction sampling for zero-order optimization, dramatically reducing memory requirements for fine-tuning large language models without sacrificing performance.