·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Towards Speed-of-Light Text Generation with Nemotron-Labs Diffusion Language Models7h◆AI is being used to resurrect the voices of dead pilots8h◆Google goes for the glitter with disco-ball icons: ‘Are y’all sure you still want this?’10h◆How VCs and founders use inflated ‘ARR’ to crown AI startups11h◆Google’s AI search is so broken it can ‘disregard’ what you’re looking for11h◆Elon Musk can’t hear you over the sound of his $1.75 trillion IPO13h◆Catch up on the Dialogues stage at Google I/O 2026.14h◆Elon, stop trying to make Grok happen14h◆You can no longer Google the word ‘disregard’15h◆We tried Google’s AI glasses and they’re almost there16h◆Specialization Beats Scale: A Strategic Variable Most AI Procurement Decisions Overlook16h◆SpaceX files to go public, and the math requires a little faith17h◆The literary world isn’t prepared for AI17h◆Spotify says its AI remix tool is for superfans, but I’m not convinced17h◆Samsung’s memory chip employees negotiated $340,000 bonuses this year20h◆Google I/O showed how the path for AI-driven science is shifting22h◆MTR-Bench: A Comprehensive Benchmark for Multi-Turn Reasoning Evaluation1d◆Fine-grained Claim-level RAG Benchmark for Law1d◆Fix the Structural Bottleneck: Context Compression via Explicit Information Transmission1d◆Linear Dynamics in the RLVR Training of Large Language Models1d◆Towards Speed-of-Light Text Generation with Nemotron-Labs Diffusion Language Models7h◆AI is being used to resurrect the voices of dead pilots8h◆Google goes for the glitter with disco-ball icons: ‘Are y’all sure you still want this?’10h◆How VCs and founders use inflated ‘ARR’ to crown AI startups11h◆Google’s AI search is so broken it can ‘disregard’ what you’re looking for11h◆Elon Musk can’t hear you over the sound of his $1.75 trillion IPO13h◆Catch up on the Dialogues stage at Google I/O 2026.14h◆Elon, stop trying to make Grok happen14h◆You can no longer Google the word ‘disregard’15h◆We tried Google’s AI glasses and they’re almost there16h◆Specialization Beats Scale: A Strategic Variable Most AI Procurement Decisions Overlook16h◆SpaceX files to go public, and the math requires a little faith17h◆The literary world isn’t prepared for AI17h◆Spotify says its AI remix tool is for superfans, but I’m not convinced17h◆Samsung’s memory chip employees negotiated $340,000 bonuses this year20h◆Google I/O showed how the path for AI-driven science is shifting22h◆MTR-Bench: A Comprehensive Benchmark for Multi-Turn Reasoning Evaluation1d◆Fine-grained Claim-level RAG Benchmark for Law1d◆Fix the Structural Bottleneck: Context Compression via Explicit Information Transmission1d◆Linear Dynamics in the RLVR Training of Large Language Models1d◆
News/CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves
arxiv
PublishedMay 19, 2026 at 4:00 AM
—neutral

CurveBench: A Benchmark for Exact Topological Reasoning over Nested Jordan Curves

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

arXiv:2605.14068v2 Announce Type: replace-cross Abstract: We introduce CurveBench, a benchmark for hierarchical topological reasoning from visual input. CurveBench consists of \textbf{756 images} of pairwise non-intersecting Jordan curves across easy, polygonal, topographic-inspired, maze-like, and

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
05
  • 01
    Gemini 3.1 Pro
  • 02
    Qwen3-VL-8B
  • 03
    Qwen-3-VL-8B-Thinking
  • 04
    GPT-5.4
  • 05
    Claude Opus 4.5
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#benchmark#computer vision#topological reasoning#visual input

No replies yet. Be first.

Mentioned models
05
  • 01
    Gemini 3.1 Pro
  • 02
    Qwen3-VL-8B
  • 03
    Qwen-3-VL-8B-Thinking
  • 04
    GPT-5.4
  • 05
    Claude Opus 4.5
Source
↗
arxiv
Read original ↗All from arxiv →
Tags
04
#benchmark#computer vision#topological reasoning#visual input

Related coverage

More from ARXIV
arxivMTR-Bench: A Comprehensive Benchmark for Multi-Turn Reasoning Evaluation1darxivFine-grained Claim-level RAG Benchmark for Law1darxivFix the Structural Bottleneck: Context Compression via Explicit Information Transmission1darxivLinear Dynamics in the RLVR Training of Large Language Models1d
The Bubble Brief
WEEKLY

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

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

Originally published on arxiv ↗
HomeModelsNews