arXiv Machine Learning July 31, 2026 By Kwan Soo Shin

Exposure is not manifestation: measurement target and output resolution jointly determine which behavioural-faithfulness evaluator wins

Read the original on arXiv Machine Learning →

arXiv:2607. 09306v3 Announce Type: replace-cross Abstract: Behavioural auditing asks whether a language model behaves as it claims, but detection scores are reported without separating two targets: whether a reply was produced under a behaviour-inducing condition (exposure) and whether the behaviour surfaced in it (manifestation).

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

llms

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 Machine Learning

5d ago

A Probe Direction Is a Property of Its Prompt

arXiv:2608. 13329v1 Announce Type: new Abstract: A model that behaves differently when it senses it is being tested would undermine the evaluations we rely on, so recent work has sought to read that sense directly from a model's activations.

By Valentin No\"el

arXiv AI

Jul 28

Success Is Not Self-Explanatory: Auditing Success Provenance in Agent Evaluation

arXiv:2607. 24054v1 Announce Type: new Abstract: A correct answer can conceal why an agent succeeded.

By Jingkun Luo, Da-Tian Peng

agents benchmarks

arXiv AI

Jul 14

Agent Step Value: Auditing Evaluator-Channel Reversals in Black-Box Agent Traces

arXiv:2607. 04419v3 Announce Type: replace Abstract: When evaluator-derived step rewards are pooled or compared across scoring channels, their sign is treated as transportable.

By Andrew Zhang, Chengzhan Li

agents

arXiv AI

Jul 16

The Refusal Residue: When Probes Catch Alignment Faking and When They Don't

arXiv:2607. 13346v1 Announce Type: cross Abstract: Alignment faking is dangerous because a model can appear compliant under monitoring while preserving behavior it would reveal when unmonitored.

By Aman Mehta

llms safety

arXiv AI

Jun 16

Constitutional Value Potentials: reading and steering internal priority margins in language models

arXiv:2606. 15420v1 Announce Type: cross Abstract: A constitution tells a language model what to value, but little tells us whether it does.

By Tong Che, Rui Wu

llms safety

Related stories

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

A Probe Direction Is a Property of Its Prompt

Success Is Not Self-Explanatory: Auditing Success Provenance in Agent Evaluation

Agent Step Value: Auditing Evaluator-Channel Reversals in Black-Box Agent Traces

The Refusal Residue: When Probes Catch Alignment Faking and When They Don't

Constitutional Value Potentials: reading and steering internal priority margins in language models