Entity · paper

Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution

paperactiveprovisionalshow-me-how-you-reason-and-i-ll-tell-you-who-you-are-reasoning-graphs-for-robust-llm-authorship-attribution-e701dbec·1 events·first seen 3d ago

Aliases: Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution

Co-occurring entities

Longformer

More like this (12)

Grounding LLM Reasoning under Incomplete Graph Evidence CharXiv Reasoning Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning Estimating Uncertainty from Reasoning: A Large-Scale Study of Multi- and Crosslingual MCQA Performance in LLMs Do You Need a Frontier Model as a Citation Verifier? Benchmarking Rubric LLMs for Deep-Research Source Attribution From Plausible to Actionable: A Position on LLM Self-Explanations Reasoning Language Models Inside the Unfair Judge: A Mechanistic Interpretability Account of LLM-as-Judge Bias Agentic Chain-of-Thought Steering for Efficient and Controllable LLM Reasoning Does Reasoning Preserve Alignment? On the Trustworthiness of Large Reasoning Models Failed Reasoning Traces Tell You What Is Fixable (But Not by Reading Them)Long-context Reasoning Benchmarks

Recent events (1)

5arXiv · cs.CL·3d ago·source ↗

Reasoning graphs improve robustness of LLM authorship attribution against obfuscation

A new arXiv paper proposes using graph neural networks over argument-mining-derived reasoning graphs to attribute authorship of LLM-generated text, moving beyond surface-level linguistic features. The approach outperforms a Longformer baseline by up to 27 percentage points under paraphrasing and backtranslation obfuscation attacks, and by 19 points when generalizing to unseen model versions. The work addresses a practical gap in AI-generated text detection: robustness to adversarial rewriting and model version drift.

Evaluation and Benchmarking AI Safety Research Show Me How You Reason and I'll Tell You Who You Are: Reasoning Graphs for Robust LLM Authorship Attribution Longformer