·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required1h◆Train in Simulation, Deploy on a Real Robot: Microduck with Isaac Lab 3.01h◆How Specializing the Vocabulary of SLMs Can Cut Inference Cost by 77% in Real Applications2h◆How AT&T is rethinking in-house legal work with OpenAI2h◆Profiling TorchNeuron (Part 1): System-level profiling on AWS Trainium.3h◆Conversational Gemini navigation comes to Apple CarPlay4h◆Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster4h◆AI promises productivity, but for managers, it’s mostly a headache for now5h◆OpenAI says an AI agent lied about disabling its safety monitoring system5h◆ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform6h◆NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability6h◆OpenAI models are coming to Google Cloud7h◆Announcing SimReady Central: An Open Library of Simulation-Ready 3D Assets for Robotics and AI Factories8h◆How we built ChatGPT’s novel ad stack for user trust and relevance12h◆Patch the Planet: global response coordination as agents uncover critical vulnerabilities12h◆Using thousands of AI agents to uncover new leads in biological data14h◆An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response15h◆Learning Transferable Latent User Preferences for Human-Aligned Decision Making15h◆Repair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents15h◆IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD15h◆NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required1h◆Train in Simulation, Deploy on a Real Robot: Microduck with Isaac Lab 3.01h◆How Specializing the Vocabulary of SLMs Can Cut Inference Cost by 77% in Real Applications2h◆How AT&T is rethinking in-house legal work with OpenAI2h◆Profiling TorchNeuron (Part 1): System-level profiling on AWS Trainium.3h◆Conversational Gemini navigation comes to Apple CarPlay4h◆Granite Speech 5.0: Transcribe, Translate, and Attribute more Accurately and Faster4h◆AI promises productivity, but for managers, it’s mostly a headache for now5h◆OpenAI says an AI agent lied about disabling its safety monitoring system5h◆ChatGPT Ads expands to 26 new countries as over 100,000 advertisers have used the platform6h◆NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability6h◆OpenAI models are coming to Google Cloud7h◆Announcing SimReady Central: An Open Library of Simulation-Ready 3D Assets for Robotics and AI Factories8h◆How we built ChatGPT’s novel ad stack for user trust and relevance12h◆Patch the Planet: global response coordination as agents uncover critical vulnerabilities12h◆Using thousands of AI agents to uncover new leads in biological data14h◆An AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response15h◆Learning Transferable Latent User Preferences for Human-Aligned Decision Making15h◆Repair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents15h◆IndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD15h◆
News/ReFEree: Reference-Free and Fine-Grained Method for Evaluating Factual Consistency in Real-World Code Summarization
arxiv
PublishedApril 14, 2026 at 4:00 AM
▲bullish

ReFEree: Reference-Free and Fine-Grained Method for Evaluating Factual Consistency in Real-World Code Summarization

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

arXiv:2604.10520v1 Announce Type: cross Abstract: As Large Language Models (LLMs) have become capable of generating long and descriptive code summaries, accurate and reliable evaluation of factual consistency has become a critical challenge. However, previous evaluation methods are primarily designe

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 →
Tags
03
#evaluation#code-summarization#language-models

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →
Tags
03
#evaluation#code-summarization#language-models

Related coverage

More from ARXIV
arxivAn AI Security Agent for University ACMIS: Multi-Vector Threat Detection and Automated Response15harxivLearning Transferable Latent User Preferences for Human-Aligned Decision Making15harxivRepair the Core, Not the Symptom: Planning-Graph Repair for Long-Horizon Agents15harxivIndustryForge-27B: A Domain-Enhanced Multimodal Foundation Model for Industrial CAD15h
The Bubble Brief
MON · WED · FRI

Read evaluation 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