·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Why the first GPU financiers are turning to inference chips in a $400 million deal3h◆A scorecard for the AI age5h◆The risk of weather data sabotage is rising6h◆Capability from Access Structure, Not Scale: Lower Bounds and Pre-Registered Tests for Hybrid Sequence Models11h◆The Steering Budget: Examples beat Knobs11h◆Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation11h◆Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions11h◆Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstration11h◆Just Keep Prompting: Evaluating Repetitive Socratic Prompting in VLMs11h◆LBA: Textual Hard-Label Adversarial Attack under Low Query Budgets11h◆Automatically Evolving Prompt Guidelines for Task-Specific Optimization11h◆Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs11h◆Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool Utility11h◆Simplicity Paradox: Debunking myths about prompting and datasets for LLM evaluation11h◆MAPS: Modeling Co-Existing Subjective Perspectives and Shared Meaning in Multi-Agent Cognitive Dialogue11h◆Introspection Fine-Tuning (IFT): Training Small LLMs to Introspect11h◆Information-Theoretic Limits of Reliability and Scaling in Language Models11h◆T5-CSBoost: Adversarial Perturbation Resistant LLM Fingerprinting11h◆Semantic Audio-driven Understanding for Dynamic Humanoid Whole Body Control11h◆LIGO-PINN: Learned Initialization via Gated Optimization to Alleviate Convergence Failures in Physics Informed Neural Networks11h◆Why the first GPU financiers are turning to inference chips in a $400 million deal3h◆A scorecard for the AI age5h◆The risk of weather data sabotage is rising6h◆Capability from Access Structure, Not Scale: Lower Bounds and Pre-Registered Tests for Hybrid Sequence Models11h◆The Steering Budget: Examples beat Knobs11h◆Automatic Hard Example Synthesis with Multi-Level Agentic Data Curation11h◆Tracing LLM Behavior to the Training Data with Empirical Next-Token Distributions11h◆Instrument Effects in Language-Model Honesty Evaluation: An Auditable Single-System Demonstration11h◆Just Keep Prompting: Evaluating Repetitive Socratic Prompting in VLMs11h◆LBA: Textual Hard-Label Adversarial Attack under Low Query Budgets11h◆Automatically Evolving Prompt Guidelines for Task-Specific Optimization11h◆Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs11h◆Eta Given Delta: Defining LLM Tool Efficiency With Marginal Tool Utility11h◆Simplicity Paradox: Debunking myths about prompting and datasets for LLM evaluation11h◆MAPS: Modeling Co-Existing Subjective Perspectives and Shared Meaning in Multi-Agent Cognitive Dialogue11h◆Introspection Fine-Tuning (IFT): Training Small LLMs to Introspect11h◆Information-Theoretic Limits of Reliability and Scaling in Language Models11h◆T5-CSBoost: Adversarial Perturbation Resistant LLM Fingerprinting11h◆Semantic Audio-driven Understanding for Dynamic Humanoid Whole Body Control11h◆LIGO-PINN: Learned Initialization via Gated Optimization to Alleviate Convergence Failures in Physics Informed Neural Networks11h◆
News/Paraphrase Brittleness in Production Retrieval-Augmented Commercial Recommendation: Reproducibility Below the Rerun-Stability Baseline
arxiv
PublishedMay 28, 2026 at 4:00 AM

Paraphrase Brittleness in Production Retrieval-Augmented Commercial Recommendation: Reproducibility Below the Rerun-Stability Baseline

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

arXiv:2605.27440v1 Announce Type: cross Abstract: Small changes to how a buyer phrases a question -- "best CRM" vs "top CRM" vs "best CRM for a SaaS startup" -- produce substantially different brand recommendations from AI assistants. Across ~6,000 paraphrase runs and ~6,000 same-prompt rerun contro

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
arxivCapability from Access Structure, Not Scale: Lower Bounds and Pre-Registered Tests for Hybrid Sequence Models11harxivThe Steering Budget: Examples beat Knobs11harxivAutomatic Hard Example Synthesis with Multi-Level Agentic Data Curation11harxivTracing LLM Behavior to the Training Data with Empirical Next-Token Distributions11h
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