AI
When Uncertainty Isn't Enough: An Empirical Study of Self-Correction in Code Generation
Researchers studied whether uncertainty estimation methods developed for natural language processing...
AI
Cross-Domain Industrial Fault Detection by Causal Mechanism Monitoring
Researchers have proposed a new method for detecting faults in industrial systems. The current appro...
AI
Position: AI Agents in Scientific Teams Should Be Studied as Human-Agent Systems
Researchers argue that AI agents in scientific teams should be studied as human-agent systems (HAS),...
AI
Beyond Correctness: Toward Automated Novelty Verification with Lean 4
Researchers have developed a system called AViD Journal that can verify whether an automatically gen...
AI
Auditing an AI-Generated Mathematical Proof: A Correction to a Greedy Conditioning Lemma in Quantum Parallel Repetition
A mathematical proof in OpenAI's 'Ten Advances in Mathematics and Theoretical Co...
AI
When Agentic Executions Fail: Detecting and Localizing Runtime Faults from Telemetry
Researchers have developed a benchmark for detecting and localizing runtime faul...
AI
A Comprehensive Survey of Wireless Foundation Models for AI-Native 6G Networks
Researchers have published a comprehensive survey of wireless foundation models ...
AI
Synchronized Logit Steering: Real-world Steganography
Researchers have developed a new method for embedding hidden messages in large l...