·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Former Plaud exec raises $26M for AI app that listens to you in the background17m◆SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices9h◆VETR: Execution-Grounded Self-Improving Agent for Symbolic Graphics Reasoning9h◆AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware9h◆SEEDS: Preview-Guided Evidence Admission for Search Agents9h◆OccludeBench: Can VLMs Find What's Missing?9h◆Towards Easier-to-Tune Training Algorithms: New Results from AlgoPerf9h◆StereoEngine: Scaling Stereo Bimanual Robot Data without Stereo Cameras9h◆Dissecting Representation Structure in Vision Transformers: A Rigorous Architectural Study9h◆Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding9h◆Forecasting Corpus-Sampling Disagreement in Masked Text Reconstruction9h◆Probabilistic Calibration Is a Trainable Capability in Language Models9h◆A Critical Look at Power Sampling when Test Time Sharpening LLMs9h◆CoBe: A Benchmark For Conversational Counterfactual Text Editing9h◆Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric Policies9h◆SWE-Duel: Towards a Self-Scaling Arena for Coding Agents9h◆Exploring Attention Sinks in Text-to-Image Transformers from Diffusion and Autoregressive Models9h◆QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code9h◆OpenPA: Forced-alignment-free Multi-granular Pronunciation Assessment with Mixed-transcript Adaptation9h◆StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams9h◆Former Plaud exec raises $26M for AI app that listens to you in the background17m◆SmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices9h◆VETR: Execution-Grounded Self-Improving Agent for Symbolic Graphics Reasoning9h◆AI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware9h◆SEEDS: Preview-Guided Evidence Admission for Search Agents9h◆OccludeBench: Can VLMs Find What's Missing?9h◆Towards Easier-to-Tune Training Algorithms: New Results from AlgoPerf9h◆StereoEngine: Scaling Stereo Bimanual Robot Data without Stereo Cameras9h◆Dissecting Representation Structure in Vision Transformers: A Rigorous Architectural Study9h◆Surfing the Activation Wave: Analyzing the Information and Geometry of LLM Hidden States for Speculative Decoding9h◆Forecasting Corpus-Sampling Disagreement in Masked Text Reconstruction9h◆Probabilistic Calibration Is a Trainable Capability in Language Models9h◆A Critical Look at Power Sampling when Test Time Sharpening LLMs9h◆CoBe: A Benchmark For Conversational Counterfactual Text Editing9h◆Beyond State-Wise Mirror Descent: Offline Policy Optimization with Parametric Policies9h◆SWE-Duel: Towards a Self-Scaling Arena for Coding Agents9h◆Exploring Attention Sinks in Text-to-Image Transformers from Diffusion and Autoregressive Models9h◆QTestBench: A Test Framework for Evaluating LLM-Generated Quantum Code9h◆OpenPA: Forced-alignment-free Multi-granular Pronunciation Assessment with Mixed-transcript Adaptation9h◆StreamDecisionBench: Evaluating Decisions in Force on Evolving Language Streams9h◆
News/From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale
arxiv
PublishedOctober 12, 2026 at 4:00 AM

From 2.1M Hours of Raw Broadcast Audio to Finnish Speech SSL: Confidence Filtering and Architectural Analysis at Scale

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

arXiv:2610.13102v1 Announce Type: new Abstract: Broadcast archives offer scale and diversity for self-supervised learning (SSL) but contain substantial noisy, non-speech, and multilingual content. Prior work typically relies on voice activity detection or language identification (LID) to extract spe

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 →

No replies yet. Be first.

Source
↗
arxiv
Read original ↗All from arxiv →

Related coverage

More from ARXIV
arxivSmolWAM: A Latent World-Action Model for Robotic Manipulation on Edge Devices9harxivVETR: Execution-Grounded Self-Improving Agent for Symbolic Graphics Reasoning9harxivAI-Based On-Board Maritime Object Detection for Earth Observation Payload Data Reduction on Versal Embedded Hardware9harxivSEEDS: Preview-Guided Evidence Admission for Search Agents9h
The Bubble Brief
MON · WED · FRI

Read AI 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