paper-with-me

Papers

Training Leaves Traces: Centered Residual Signatures for Language Model Lineage Verification

2026-08-14 · Aman Singh Thakur, Rayan Khoury hf

Open-weight language models are fine-tuned, quantized, pruned, and merged, yet their provenance is often undocumented. We study data-free white-box lineage verification: can weights alone reveal whether two compatible model checkpoints share ancestry? Residual training produces a shared identity-aligned component in branch products, so this structure alone cannot establish ancestry. We remove it and compare checkpoint-specific structure across residual blocks, yielding a symmetric lineage score calibrated against independent checkpoints. On residual-MLP and GPT-2 benchmarks, the score separates fine-tuned, LoRA-merged, pruned, and quantized descendants from independent and distilled models (AUROC=1.0), distinguishing weight ancestry from behavioral similarity. Under function-preserving checkpoint laundering experiments, weight-space baselines lose margin or fail; our score remains unchanged and runs 76x faster than the nearest robust baseline on GPT-2. The projection-pairing signal appears across six language-model families and beyond, and a case study correctly identifies 3 related and 7 unrelated LLaMA-2 public checkpoints. Collectively, these results establish a passive, data-free provenance signal for compatible open-weight language-model checkpoints

📄 PDF Abstract BibTeX arXiv:2608.14929

Code (0)

등록된 구현이 없습니다.

Similar Papers 제목 키워드 기반

Unlearning Isn't Invisible: Detecting Unlearning Traces in LLMs from Model Outputs

2025-06-16 · YiWei Chen, Soumyadeep Pal, Yimeng Zhang, Qing Qu 외

Machine unlearning (MU) for large language models (LLMs), commonly referred to as LLM unlearning, seeks to remove specific undesirable data or knowledge from a trained model, while maintaining its performance on standard…

Machine Unlearning

Internal Flow Signatures for Self-Checking and Refinement in LLMs

2026-02-02 · Sungheon Jeong, Sanggeon Yun, Ryozo Masukawa, Wenjun Haung 외 arxiv

Large language models can generate fluent answers that are unfaithful to the provided context, while many safeguards rely on external verification or a separate judge after generation. We introduce \emph{internal flow si…

Deep Fake Detection: Survey of Facial Manipulation Detection Solutions

2021-06-23 · Samay Pashine, Sagar Mandiya, Praveen Gupta, Rashid Sheikh

Deep Learning as a field has been successfully used to solve a plethora of complex problems, the likes of which we could not have imagined a few decades back. But as many benefits as it brings, there are still ways in wh…

Survey

Do Vision Models Truly Forget? New Findings from Representation-Level Certification of Visual Unlearning in Vertical Federated Learning

2026-05-19 · Zhenyu Yu, Yangchen Zeng, Chunlei Meng, Guangzhen Yao 외 arxiv

Machine unlearning in Vertical Federated Learning (VFL) has attracted growing interest, yet existing methods certify forgetting solely using output-level metrics. We challenge these works by introducing Mirage, a represe…

Federated Learning

On the Invariants of Softmax Attention

2026-04-06 · Wonsuk Lee arxiv

Softmax attention maps every query--key interaction into a probability distribution, but the underlying structure remains largely unexplored. We define the \emph{energy field}, the row-centered attention logit, and show …