The paper introduces a hybrid system, HYBRIDSOURCETRACKER (HST), that combines vector search and Winnowing fingerprinting to achieve scalable, high-precision provenance tracking for code generated by…

The paper proposes interaction-layer antidistillation watermarks by embedding behavioral markers into the system prompt, which successfully track knowledge distillation even when paraphrasing attacker…

This paper investigates the vulnerability of LLM-based automatic grading systems to prompt injection (PI) attacks, demonstrating that current systems are highly susceptible to manipulation that can le…

The paper proposes SAFESEAL, a novel key-conditioned watermarking framework that embeds robust, provider-specific watermarks into LLM outputs with minimal semantic distortion, effectively protecting i…

PASA introduces a robust, semantic-level watermarking technique that embeds and detects watermarks in the latent embedding space, successfully resisting semantic-invariant attacks like paraphrasing.

The paper introduces Synthesis Data Reversion (SDR), a method that infers the data laundering transformation used in LLM training and synthesizes queries to restore the detection signals lost when pro…

The paper introduces a distribution-free statistical framework that allows existing rewrite-based detectors to achieve finite-sample False Discovery Rate (FDR) guarantees for detecting LLM-generated t…

The paper introduces a novel framework using steganographic canary files to detect and block unauthorized processing of sensitive documents by LLMs, even when the data passes through traditional secur…

The paper demonstrates that the AI-like style introduced by post-training alignment can be measured, localized, and causally removed using a novel ablation technique called PASTA.

The paper introduces Reasoning Exposure Prompting (REP), a method that demonstrates that even when LLMs hide their internal reasoning steps from users, useful reasoning supervision can still be elicit…

The paper introduces Reasoning Exposure Prompting (REP), a method that demonstrates that even when LLMs hide internal reasoning traces from users, useful reasoning supervision can still be elicited th…

The paper introduces TSM-Bench, a new benchmark that demonstrates existing LLM-generated text detectors fail to accurately identify task-specific machine-generated content found in real-world Wikipedi…

This paper develops provably undetectable and robust watermarking schemes for LLM outputs even when the per-token entropy is only constant, removing previous dependencies on high entropy rates or larg…

The paper introduces RLSpoofer, a lightweight, black-box reinforcement learning attack that demonstrates the fragile resilience of current LLM watermarking schemes by achieving a high spoofing success…

The paper demonstrates that using raw source text for fine-tuning LLMs on vulnerability detection causes high false-positive rates by memorizing surface-level syntax, a problem mitigated by using Abst…

The paper analyzes the robustness of current LLM watermarking schemes against various text modifications, concluding that watermarks can be removed with reasonable effort.

The paper demonstrates that encoding harmful prompts as genuine mathematical problems, rather than just using mathematical formatting, effectively bypasses the safety filters of large language models.

The paper demonstrates that combining outputs from multiple large language models (LLMs) effectively cancels out statistical watermarks, revealing a fundamental vulnerability in current AI text detect…

The paper introduces the first byte-native Large Language Model (LLM) capable of analyzing raw executable binary data, achieving high accuracy in tasks like malware and architecture classification.

XMark introduces a novel multi-bit watermarking technique that reliably embeds binary messages into LLM-generated text while maintaining high text quality and robust performance even with limited toke…