Yanlin Wang

Even the top-performing LLM struggles with cross-file reasoning, achieving only 69.1% accuracy on a new benchmark designed to reflect real-world software development challenges.

PhoenixRepair redefines how software agents explore repair strategies, achieving a 76% resolution rate by leveraging multi-location sampling and iterative refinement.

Design knowledge is the game-changer that boosts repository-level code generation, leading to a 9.55 percentage point improvement in performance over existing models.

Hawk boosts NPU kernel generation accuracy by over 30% while doubling execution speed, revolutionizing how we approach hardware-specific programming.

Achieving 89.2% accuracy in microservice decomposition, MicroAgent outperforms traditional methods by a striking 24.6%.

Attackers are exploiting a critical time gap in Polymarket's architecture, allowing them to revert over 980,000 filled orders and realize profits exceeding $1.49 million.

Log-based anomaly detection models are missing 90% of the picture, but AnomalyGen uses LLMs and static analysis to hallucinate realistic training data and close the gap.

Forget scaling laws: high-quality training data lets smaller LLMs crush larger ones at agentic coding tasks.