·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Closing the AI guidance gap in education10m◆Otter AI can now follow conversations across 30+ languages2h◆Phantom Consensus: Measuring Social Reasoning in Multi-Agent LLM Decision-Making4h◆ROSE: Risk-Aware Orthogonal Subspace Navigation for Lifelong Knowledge Editing in Multimodal Large Language Models4h◆Measuring and Mitigating the Distributional Gap Between Real and Simulated User Behaviors4h◆CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs4h◆FindLink: Learning to Assess Scientific Connections from Source-Grounded Premises4h◆Prompt Lottery in LLM-as-a-Judge: Measuring and Mitigating Sensitivity to Syntactic Rephrasing4h◆AlienBench: A Benchmark of Structural Novelty in Language Model Outputs4h◆Internal Safety Collapse in LLMs and Coding Agents4h◆Semantic Conditional Similarity (SCS): A Formal Mathematical Framework4h◆Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters4h◆Copies Are Not Confirmations: How News Duplication Fools LLM Judges and Continual Learners4h◆Safety and accuracy diverge across deployment conditions in clinical large language models4h◆DunbaaBERT: From Sacrifice to Semantics4h◆Sampling Reveals Style: Unsupervised, Training-Free Discovery of Prompt-Conditional Stylistic Axes in LLM Activations4h◆Measuring Form and Function in Language Models4h◆Does Reasoning Improve Psychological Depth in Large Language Models?4h◆Rethinking Data Selection as Distribution Matching on Semantic Manifolds4h◆AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research4h◆Closing the AI guidance gap in education10m◆Otter AI can now follow conversations across 30+ languages2h◆Phantom Consensus: Measuring Social Reasoning in Multi-Agent LLM Decision-Making4h◆ROSE: Risk-Aware Orthogonal Subspace Navigation for Lifelong Knowledge Editing in Multimodal Large Language Models4h◆Measuring and Mitigating the Distributional Gap Between Real and Simulated User Behaviors4h◆CodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs4h◆FindLink: Learning to Assess Scientific Connections from Source-Grounded Premises4h◆Prompt Lottery in LLM-as-a-Judge: Measuring and Mitigating Sensitivity to Syntactic Rephrasing4h◆AlienBench: A Benchmark of Structural Novelty in Language Model Outputs4h◆Internal Safety Collapse in LLMs and Coding Agents4h◆Semantic Conditional Similarity (SCS): A Formal Mathematical Framework4h◆Steerable Multi-Objective Group-Relative Preference Optimization with Jointly Trained Low-Rank Adapters4h◆Copies Are Not Confirmations: How News Duplication Fools LLM Judges and Continual Learners4h◆Safety and accuracy diverge across deployment conditions in clinical large language models4h◆DunbaaBERT: From Sacrifice to Semantics4h◆Sampling Reveals Style: Unsupervised, Training-Free Discovery of Prompt-Conditional Stylistic Axes in LLM Activations4h◆Measuring Form and Function in Language Models4h◆Does Reasoning Improve Psychological Depth in Large Language Models?4h◆Rethinking Data Selection as Distribution Matching on Semantic Manifolds4h◆AblationBench: Evaluating Automated Planning of Ablations in Empirical AI Research4h◆
News/Guardrail Erosion in Domain Fine-Tuned Large Language Models: A Paired Audit of Knowledge-Centric Abstention in Biomedical LLMs
arxiv
PublishedNovember 10, 2026 at 5:00 AM

Guardrail Erosion in Domain Fine-Tuned Large Language Models: A Paired Audit of Knowledge-Centric Abstention in Biomedical LLMs

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

arXiv:2611.08780v1 Announce Type: new Abstract: The number of domain fine-tuned models, particularly in the biomedical domain, is rapidly expanding. Fine-tuning promises high-performing models at a fraction of the cost of training a base model, but it can also produce unintended behavioral changes.

Stay posted· Newsletter

Three short emails a week: Monday, Wednesday and Friday. Top movers, new models and the most-referenced story.

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

Your email is used only to send The Bubble Brief. Privacy policy

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
arxivPhantom Consensus: Measuring Social Reasoning in Multi-Agent LLM Decision-Making4harxivROSE: Risk-Aware Orthogonal Subspace Navigation for Lifelong Knowledge Editing in Multimodal Large Language Models4harxivMeasuring and Mitigating the Distributional Gap Between Real and Simulated User Behaviors4harxivCodeQuant2: Enhanced Outlier Mitigation and Adaptive Clustering for Low-Precision LRMs4h
The Bubble Brief
MON · WED · FRI

Read AI insights every Monday, Wednesday and Friday — top movers, new models, story of the week.

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

Your email is used only to send The Bubble Brief. Privacy policy

Originally published on arxiv ↗
PrivacyTermsRefundsLegal notice|AboutMethodologyContact

© 2026 Databubble

Built by Marouane Gazouzi · Readyo, AI interview coach

HomeModelsNews