PulseAugur
EN
LIVE 20:01:10

New framework detects AI-generated code at token, line, and block levels

By PulseAugur Editorial · [2 sources] ·

Researchers have developed a new framework to detect and localize AI-generated code within files. This method models AI authorship at the token level, using a code language model with low-rank adapters to analyze per-token log-likelihood ratios. The system then aggregates this evidence to predict AI involvement at the line, block, and file levels, offering a more granular approach than existing methods that provide only a single score per file. The framework achieved a 72.4 macro F1 score on an internal test set and demonstrated a low false-positive rate on pre-2022 human-written code. AI

IMPACT This framework could improve code integrity and security by accurately identifying AI-generated code segments.

RANK_REASON This is a research paper detailing a new framework for code analysis.

Read on Hugging Face Daily Papers →

AI-generated summary · Google Gemini · from 2 sources. How we write summaries →

New framework detects AI-generated code at token, line, and block levels

How we ranked this

Signal score
0 / 100
Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket
Research
This is a research paper detailing a new framework for code analysis.
Source corroboration
2 independent sources
Multiple independent publishers reporting the same story raises confidence that it's real and newsworthy.
Topics
paper, product
Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance
High
Clearly on-topic for AI-industry coverage.
Story freshness
21 days old
Aged out of breaking-news scoring windows; ranking reflects the durable signal from the full source set.

Full methodology in our editorial standards.

COVERAGE [2]

  1. arXiv cs.CL TIER_1 English(EN) · Yue Xu, Xiyang Wang, Hui Liu, Bolei He ·

    From Tokens to Files: Multi-Granularity Detection and Localization of AI-Involved Code

    arXiv:2610.27863v1 Announce Type: cross Abstract: The increasing use of large language models for code generation calls for methods that distinguish purely human-written code from AI-involved code: files in which any part, however small, was generated or substantially modified by…

  2. Hugging Face Daily Papers TIER_1 English(EN) ·

    From Tokens to Files: Multi-Granularity Detection and Localization of AI-Involved Code

    The increasing use of large language models for code generation calls for methods that distinguish purely human-written code from AI-involved code: files in which any part, however small, was generated or substantially modified by a model. Beyond flagging such files, we aim to lo…

RELATED ENTITIES

RELATED TOPICS