·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
AI has a water problem. Google thinks it has a fix2h◆Google must let publishers opt out of AI Search features, rules UK3h◆FederatedSkill: Federated Learning for Agentic Skill Evolution7h◆Toward a Modular Architecture for Embedded AI Agent Systems at the Edge7h◆A Graph Foundation Model with Spectral Parsing and Prototype-Guided Spatial Propagation7h◆Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate7h◆Evaluating the Reversal Curse in Model Editing7h◆Fast Unlearning at Scale via Margin Self-Correction7h◆Can Local Learning Match Self-Supervised Backpropagation?7h◆CAPER: Clause-Aligned Process Supervision for Text-to-SQL7h◆An Asymptotic Theory of Chain-of-Thought in In-Context Learning7h◆DeskCraft: Benchmarking Desktop Agents on Professional Workflows and Human-in-the-Loop Collaboration7h◆Enhanced Renewable Energy Forecasting using Context-Aware Conformal Prediction7h◆Don't Gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems7h◆RMPrior: Bridging Propagation Priors and Diffusion Refinement for Efficient Radio Map Construction7h◆Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR7h◆RogueMerge: Robust and Unified Attacks against LLM Model Merging7h◆The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset7h◆MLSkip: Data Skipping for ML Filters via Lightweight Metadata7h◆Cross-Modal Contrastive Learning of ECG and Angiography Representations for Severe Stenosis Classification7h◆AI has a water problem. Google thinks it has a fix2h◆Google must let publishers opt out of AI Search features, rules UK3h◆FederatedSkill: Federated Learning for Agentic Skill Evolution7h◆Toward a Modular Architecture for Embedded AI Agent Systems at the Edge7h◆A Graph Foundation Model with Spectral Parsing and Prototype-Guided Spatial Propagation7h◆Anomalies in Multivariate Time Series Benchmarks Are Mostly Univariate7h◆Evaluating the Reversal Curse in Model Editing7h◆Fast Unlearning at Scale via Margin Self-Correction7h◆Can Local Learning Match Self-Supervised Backpropagation?7h◆CAPER: Clause-Aligned Process Supervision for Text-to-SQL7h◆An Asymptotic Theory of Chain-of-Thought in In-Context Learning7h◆DeskCraft: Benchmarking Desktop Agents on Professional Workflows and Human-in-the-Loop Collaboration7h◆Enhanced Renewable Energy Forecasting using Context-Aware Conformal Prediction7h◆Don't Gamble, GAMBLe: An Analytical Framework for AI-Driven Research Systems7h◆RMPrior: Bridging Propagation Priors and Diffusion Refinement for Efficient Radio Map Construction7h◆Learning to Solve, Forgetting to Retain: Correct-Set Turnover in RLVR7h◆RogueMerge: Robust and Unified Attacks against LLM Model Merging7h◆The Road Ahead in Autonomous Driving: The KITScenes Multimodal Dataset7h◆MLSkip: Data Skipping for ML Filters via Lightweight Metadata7h◆Cross-Modal Contrastive Learning of ECG and Angiography Representations for Severe Stenosis Classification7h◆
News/A Paired Testing Protocol for Batch-Conditioned Refusal Robustness in LLM Serving
arxiv
PublishedMay 28, 2026 at 4:00 AM
—neutral

A Paired Testing Protocol for Batch-Conditioned Refusal Robustness in LLM Serving

Source
arxiv.orgfull article ↗
Read on arxiv→
Publisher summary· verbatim

arXiv:2605.27763v1 Announce Type: new Abstract: Safety evaluations of language models often treat serving configuration as fixed background infrastructure, but batch condition is an untested treatment variable whenever the same prompt may be evaluated alone, in a synchronized batch, or inside a cont

Stay posted· Newsletter

A 5-min weekly brief — top movers, price watch, story of the week.

// no spam · unsubscribe one-click · free forever

Discussion
Source
↗
arxiv
Read original ↗All from arxiv →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivFederatedSkill: Federated Learning for Agentic Skill Evolution7harxivToward a Modular Architecture for Embedded AI Agent Systems at the Edge7harxivA Graph Foundation Model with Spectral Parsing and Prototype-Guided Spatial Propagation7harxivAnomalies in Multivariate Time Series Benchmarks Are Mostly Univariate7h
The Bubble Brief
WEEKLY

Read AI insights every Tuesday — top movers, new releases, story of the week.

// no spam · unsubscribe one-click · free forever

Originally published on arxiv ↗
HomeModelsNews