AI
Symposium: Trust via Auditable Records for Communities of AI Scientist Agents
A new framework called Symposium aims to improve trust among researchers by creating auditable recor...
AI
From Retrieved Context to Runtime Control: Adaptive Compression for Edge-based RAG
Researchers propose adaptive compression techniques to optimize the performance of retrieval-augment...
AI
Enforcing LLM Safety through DMD-based Classification of Prompt-Response Embedding Dynamics
Researchers have proposed a new method to detect and prevent unsafe outputs from large language mode...
AI
Scientific Data Skills: Enabling Agent-Ready Scientific Data Services at Scale
Researchers have developed a new way to represent scientific data in a format that's usable by artif...
AI
Can Agent Memory Systems Track Evolving State?
Researchers have developed a benchmark called StateMemBench to test the ability ...
AI
Frequency-Aware Continual Learning for Smart Contract Vulnerability Detection with Large Language Models
Researchers propose a three-stage pipeline to improve smart contract vulnerabili...
AI
Learning Hierarchical Skill Policies with Offline Quality-Diversity Reinforcement Learning
Researchers propose a new approach called QDOS (Quality-Diversity Offline Skill ...
AI
Rethinking the Evaluation and Optimization of LLM-Based Social Simulation
Researchers have proposed a new evaluation and optimization method for large lan...