·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
XDOF, just three months out of stealth, is in talks for a Series B at a $1.2B valuation9h◆OpenAI’s rogue agents keep escaping, with no formal process to investigate them10h◆AI compute provider Nscale is looking for $3.5B in pre-IPO financing12h◆Architecting memory and storage in the AI era14h◆Roland is getting into generative AI music with Melody Flip15h◆What will Apple’s John Ternus era look like?16h◆Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge17h◆Microsoft says virtually nobody was grabbing NYT articles through its chatbot17h◆Apple’s Ternus era begins as Nvidia bets on the whole AI stack17h◆Google’s Gemini Spark can now manage your Google Photos library18h◆Less than 24 hours to apply for your TechCrunch Disrupt 2026 Side Event19h◆Rogue OpenAI agents appear to have organized another attack using a German wiki19h◆Instagram’s AI detection is a mess (again)21h◆Why AI food looks like that22h◆Microsoft’s Project Zenith is a ‘distraction-free Windows experience’ for developers22h◆Sam Altman apologizes for ‘messy’ GPT-6 Astra rollout that’s locked out paying users22h◆This NAS company wants to run your local smart home23h◆Data from drones in Ukraine is fueling a new Wild West marketplace1d◆The sameness problem behind those unappetizing AI-generated menus1d◆CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning1d◆XDOF, just three months out of stealth, is in talks for a Series B at a $1.2B valuation9h◆OpenAI’s rogue agents keep escaping, with no formal process to investigate them10h◆AI compute provider Nscale is looking for $3.5B in pre-IPO financing12h◆Architecting memory and storage in the AI era14h◆Roland is getting into generative AI music with Melody Flip15h◆What will Apple’s John Ternus era look like?16h◆Another swarm of OpenAI agents reached the open internet without the frontier lab’s knowledge17h◆Microsoft says virtually nobody was grabbing NYT articles through its chatbot17h◆Apple’s Ternus era begins as Nvidia bets on the whole AI stack17h◆Google’s Gemini Spark can now manage your Google Photos library18h◆Less than 24 hours to apply for your TechCrunch Disrupt 2026 Side Event19h◆Rogue OpenAI agents appear to have organized another attack using a German wiki19h◆Instagram’s AI detection is a mess (again)21h◆Why AI food looks like that22h◆Microsoft’s Project Zenith is a ‘distraction-free Windows experience’ for developers22h◆Sam Altman apologizes for ‘messy’ GPT-6 Astra rollout that’s locked out paying users22h◆This NAS company wants to run your local smart home23h◆Data from drones in Ukraine is fueling a new Wild West marketplace1d◆The sameness problem behind those unappetizing AI-generated menus1d◆CulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning1d◆
News/Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models
arxiv
PublishedJuly 18, 2026 at 4:00 AM
▲bullish

Function-Aware Fill-in-the-Middle as Mid-Training for Coding Agent Foundation Models

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

arXiv:2607.12463v2 Announce Type: replace Abstract: Coding agents must integrate external tool returns into ongoing reasoning - a capability that standard left-to-right pretraining on code exposes only in its forward direction. We observe that the action-observation-continuation loop of a coding age

Stay posted· Newsletter

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

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

Discussion
Mentioned models
02
  • 01
    Qwen2.5-Coder-Instruct
  • 02
    Qwen3-8B
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#pretraining#self-supervised#mid-training#artificial intelligence

No replies yet. Be first.

Mentioned models
02
  • 01
    Qwen2.5-Coder-Instruct
  • 02
    Qwen3-8B
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#pretraining#self-supervised#mid-training#artificial intelligence

Related coverage

More from ARXIV
arxivCulturalMenuBench: Probing the Knowledge-Application Gap in Multimodal Culinary Reasoning1d
The Bubble Brief
WEEKLY

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

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

Originally published on arxiv ↗
HomeModelsNews