·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Andrew Yang thinks the next big startup opportunity is lowering the cost of living6h◆Anthropic’s safety warnings may have just backfired — the government has pulled the plug on its most powerful AI8h◆SpaceX IPO: Live updates on everything you need to know11h◆Meta’s months-old AI unit is a soul-crushing gulag, say the engineers stuck inside it11h◆Chinese cybercrime operation that used AI to scam ‘hundreds of thousands of victims’ sued by Google14h◆Mistral is rumored to be raising €3B at €20B valuation17h◆Siri is good now??17h◆Elon Musk is the world’s first trillionaire18h◆SpaceX, Anthropic, and OpenAI’s hot IPO summer18h◆olmo-eval: An evaluation workbench for the model development loop18h◆It’s hot IPO summer, and the MANGOS are ripe19h◆SpaceX’s massive IPO: all the latest news20h◆Jeff Bezos’ AI startup aims to build an ‘artificial general engineer’20h◆New OpenAI Academy courses for the next era of work1d◆Siri won’t be your AI girlfriend1d◆Cheaper, faster, and culturally aware, Avataar’s video AI is built for India’s scale1d◆EDEN: A Large-Scale Corpus of Clinical Notes for Italian1d◆ASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection1d◆LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling1d◆APPO: Agentic Procedural Policy Optimization1d◆Andrew Yang thinks the next big startup opportunity is lowering the cost of living6h◆Anthropic’s safety warnings may have just backfired — the government has pulled the plug on its most powerful AI8h◆SpaceX IPO: Live updates on everything you need to know11h◆Meta’s months-old AI unit is a soul-crushing gulag, say the engineers stuck inside it11h◆Chinese cybercrime operation that used AI to scam ‘hundreds of thousands of victims’ sued by Google14h◆Mistral is rumored to be raising €3B at €20B valuation17h◆Siri is good now??17h◆Elon Musk is the world’s first trillionaire18h◆SpaceX, Anthropic, and OpenAI’s hot IPO summer18h◆olmo-eval: An evaluation workbench for the model development loop18h◆It’s hot IPO summer, and the MANGOS are ripe19h◆SpaceX’s massive IPO: all the latest news20h◆Jeff Bezos’ AI startup aims to build an ‘artificial general engineer’20h◆New OpenAI Academy courses for the next era of work1d◆Siri won’t be your AI girlfriend1d◆Cheaper, faster, and culturally aware, Avataar’s video AI is built for India’s scale1d◆EDEN: A Large-Scale Corpus of Clinical Notes for Italian1d◆ASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection1d◆LoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling1d◆APPO: Agentic Procedural Policy Optimization1d◆
News/Coherence Collapse: Diagnosing Why Code Agents Fail After Reaching the Right Code
arxiv
PublishedMay 28, 2026 at 4:00 AM

Coherence Collapse: Diagnosing Why Code Agents Fail After Reaching the Right Code

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

arXiv:2603.24631v2 Announce Type: replace-cross Abstract: Code agents resolve 65-70% of SWE-bench Verified issues, but Pass@1 cannot tell us why the rest fail, and, as we show, capable-model failures are systematically misdiagnosed without trajectory data. We introduce TRAJEVAL, a training-free deco

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
arxivEDEN: A Large-Scale Corpus of Clinical Notes for Italian1darxivASTER: Latent Pseudo-Anomaly Generation for Unsupervised Time-Series Anomaly Detection1darxivLoHoSearch: Benchmarking Long-Horizon Search Agents Beyond the Human Difficulty Ceiling1darxivAPPO: Agentic Procedural Policy Optimization1d
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