·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Vertu wants executives to pay $6,880 for an AI agent — here’s how it actually performs1h◆Databricks hits $188B valuation, extending its run as AI’s favorite second act2h◆The Zoom hack that says, ‘Don’t record me’3h◆Agility Robotics plants its flag in Tesla’s backyard4h◆AI-driven memory crunch jolts India’s smartphone market4h◆TikTok is testing an AI likeness detection tool4h◆How Apple’s big lawsuit could disrupt OpenAI’s IPO plans6h◆Apple’s plot to crush OpenAI6h◆Fine-tune video and image models at scale with NVIDIA NeMo Automodel and 🤗 Diffusers8h◆Patreon stops asking AI bots not to scrape — and starts blocking them9h◆Apple’s lawsuit couldn’t come at a worse time for OpenAI10h◆Why the first GPU financiers are turning to inference chips in a $400 million deal12h◆A scorecard for the AI age14h◆The risk of weather data sabotage is rising15h◆The Steering Budget: Examples beat Knobs20h◆Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs20h◆RxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination20h◆When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models20h◆SAGA: Schema-Aware Grounding for Agentic Text-to-SPARQL Generation20h◆ReasFlow: Assisting Reasoning-Centric Scientific Discovery in Applied Mathematics via a Knowledge-Based Multi-Agent System20h◆Vertu wants executives to pay $6,880 for an AI agent — here’s how it actually performs1h◆Databricks hits $188B valuation, extending its run as AI’s favorite second act2h◆The Zoom hack that says, ‘Don’t record me’3h◆Agility Robotics plants its flag in Tesla’s backyard4h◆AI-driven memory crunch jolts India’s smartphone market4h◆TikTok is testing an AI likeness detection tool4h◆How Apple’s big lawsuit could disrupt OpenAI’s IPO plans6h◆Apple’s plot to crush OpenAI6h◆Fine-tune video and image models at scale with NVIDIA NeMo Automodel and 🤗 Diffusers8h◆Patreon stops asking AI bots not to scrape — and starts blocking them9h◆Apple’s lawsuit couldn’t come at a worse time for OpenAI10h◆Why the first GPU financiers are turning to inference chips in a $400 million deal12h◆A scorecard for the AI age14h◆The risk of weather data sabotage is rising15h◆The Steering Budget: Examples beat Knobs20h◆Polestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs20h◆RxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination20h◆When a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models20h◆SAGA: Schema-Aware Grounding for Agentic Text-to-SPARQL Generation20h◆ReasFlow: Assisting Reasoning-Centric Scientific Discovery in Applied Mathematics via a Knowledge-Based Multi-Agent System20h◆
News/Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputs
arxiv
PublishedMay 12, 2026 at 4:00 AM

Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputs

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

arXiv:2605.04665v2 Announce Type: replace Abstract: When the substantive content of a request is rewritten, do large language models still answer in the format the original task asked for? We find that they often do not, even at temperature zero. On a 150-query evaluation over five compact 2025-era

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
arxivThe Steering Budget: Examples beat Knobs20harxivPolestar: Drift-Aware Cache Calibration and Token Commitment for Efficient Inference of Diffusion LLMs20harxivRxBrain: Embodied Cognition Foundation Model with Joint Language-Visual Reasoning and Imagination20harxivWhen a Verified World Model Still Loses: Play-Adequacy vs Prediction-Accuracy in LLM-Synthesized Code World Models20h
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