arXiv Machine Learning August 11, 2026 By Zhichao Fan, Yanhang Li, Zexin Zhuang, Xian Sun, Yingshuo Wang

The Moving Target: A Longitudinal Audit of Trust-Benchmark Score Drift Across Open-Source Chat LLM Release Lines

Read the original on arXiv Machine Learning →

arXiv:2607. 02587v2 Announce Type: replace-cross Abstract: Trust-benchmark scores reported on a chat-LLM release line are often carried across several checkpoints of the same line, as if the underlying model had not shifted between releases.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jul 7

The Moving Target: A Longitudinal Audit of Trustworthiness Drift Across Twelve Checkpoints of Open-Source Chat LLMs

arXiv:2607. 02587v1 Announce Type: cross Abstract: Model cards quote trust-benchmark scores without recording when they were measured, and the same number is routinely carried across successive checkpoints of one release line as if the model behind it had not shifted.

By Zhichao Fan, Yanhang Li, Zexin Zhuang, Xian Sun, Yingshuo Wang

llms benchmarks

arXiv AI

Aug 5

BulkPR-Bench: Benchmarking Queue-Level Governance of Interacting Pull Requests

arXiv:2608. 02685v1 Announce Type: cross Abstract: Coding-agent benchmarks increasingly cover long-horizon, end-to-end, and interactive development, but typically retain one requested outcome or a fixed change sequence.

By Zetong Xiong, Qiao Zhao, Jun Zhang, Xueying Lyu, Zhi Li, Yixiang Tu, Xiaowen Yang, Yunjie Zhang, Yufeng Wang, Zhe Zhang, Kaize Yu, Hanwen Du, Zhongkai Sun, Zhuoxin Liu, Zekun Lin, Jianwen Yang, Ruining Chen, Ying Zhang, Tingxuan Pan, Ke Chen, Shubin Han, Chuanhao Sun, Yehua Yang

agents benchmarks safety

arXiv Machine Learning

Jul 22

HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks

arXiv:2607. 18867v1 Announce Type: new Abstract: Large language models leak parametric knowledge of realized outcomes into historical financial decision tasks.

By Haozhe Jia

llms robotics efficiency

arXiv AI

Jun 16

Who Drifted: the System or the Judge? Anytime-Valid Attribution in LLM Evaluation Pipelines

arXiv:2606. 15474v1 Announce Type: new Abstract: Continuous evaluation of LLM products relies on a strong LLM judge treated as ground truth: a cheap monitor scores every interaction and a team is paged when the score drifts down.

By Yitao Li

llms nlp

arXiv AI

Jul 24

Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models

arXiv:2607. 20436v1 Announce Type: cross Abstract: Safety evaluations often assume that behavior observed during testing reflects behavior in ordinary use, but fine-tuning can break this assumption.

By Phongsakon Mark Konrad, Toygar Tanyel, Serkan Ayvaz

llms fine-tuning safety

arXiv AI

Jul 23

Measuring LLM Trust Allocation Across Conflicting Software Artifacts

arXiv:2604. 03447v2 Announce Type: replace-cross Abstract: LLM-based software engineering assistants often reason over multiple artifacts, including code, documentation, signatures, and tests, even when those artifacts are incomplete or mutually inconsistent.

By Noshin Ulfat, Ahsanul Ameen Sabit, Soneya Binta Hossain

llms

Related stories

The Moving Target: A Longitudinal Audit of Trustworthiness Drift Across Twelve Checkpoints of Open-Source Chat LLMs

BulkPR-Bench: Benchmarking Queue-Level Governance of Interacting Pull Requests

HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks

Who Drifted: the System or the Judge? Anytime-Valid Attribution in LLM Evaluation Pipelines

Routing Subspaces: Auditing Evaluation-to-Deployment Mismatch in Fine-Tuned Language Models

Measuring LLM Trust Allocation Across Conflicting Software Artifacts