Martin Vechev

The paper introduces SecPI, a fine-tuning pipeline that teaches reasoning language models (RLMs) to autonomously internalize structured security reasoning, significantly improving secure code generation without requiring explicit security prompts at inference.

The paper proposes a novel binomial multibit LLM watermarking scheme that encodes every bit of a payload at every token position, achieving superior message accuracy and robustness compared to existing methods.

The paper introduces Honeyval, a comprehensive evaluation framework, to rigorously test LLM-powered HTTP honeypots, demonstrating that these systems provide substantially longer and harder-to-detect interactions compared to traditional methods.

The paper introduces Honeyval, a comprehensive evaluation framework, to rigorously test LLM-powered HTTP honeypots, demonstrating that these honeypots provide substantially longer and harder-to-detect interactions compared to traditional methods.

The paper proposes using fine-grained quality signals, such as pairwise self-judgments and token-level entropy, instead of simple binary correctness to improve LLM performance on saturated datasets, showing significant gains on simple tasks but requiring careful calibration for complex ones.