·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Meta-ethics and AI: exploring the novel meta-ethical questions in the era of AI8h◆SSAKG 2.0: An Open-Source Package for Structural Associative Sequence Memory and Context-Based Retrieval8h◆Epistemic Sybil Resistance: Multiplying AI Agents Without Multiplying Evidence8h◆DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents8h◆MASkills: Continual Skills Optimization for Multi-Agent LLM Systems8h◆Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model8h◆SpecMine: A Large-Scale Corpus of Spec-Driven Development Artifacts8h◆Elite political incivility is rising across democracies8h◆Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation8h◆Sim2Signal: Sim-to-Real Benchmarks for Traffic Signal Control8h◆Recursive Value Learning for Long-Horizon Offline Goal-Conditioned RL8h◆Similarity-Aware Personalized Federated Learning in Heterogeneous Environments8h◆Oracle, will I ever learn? A study of prediction convergence and complementarity across link prediction models8h◆UE5M3 FP4 Block Scaling for Stable Language Model Pretraining8h◆Network-Aware Forecasting on Wireless Access Points8h◆Monotonic anomaly detection8h◆Cantelli Constrained Policy Optimization8h◆Secure AI-Driven Super-Resolution for Real-Time Mixed Reality Applications8h◆From Digital to Physical Reservoir Computing: Co-Optimizing Soft Robotic Reservoirs via Dynamics Matching8h◆Optimal Transport for Network Comparison: A Review with Machine Learning Applications8h◆Meta-ethics and AI: exploring the novel meta-ethical questions in the era of AI8h◆SSAKG 2.0: An Open-Source Package for Structural Associative Sequence Memory and Context-Based Retrieval8h◆Epistemic Sybil Resistance: Multiplying AI Agents Without Multiplying Evidence8h◆DocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents8h◆MASkills: Continual Skills Optimization for Multi-Agent LLM Systems8h◆Train at Moving Edge: Online-Verified Prompt Selection for Efficient RL Training of Large Reasoning Model8h◆SpecMine: A Large-Scale Corpus of Spec-Driven Development Artifacts8h◆Elite political incivility is rising across democracies8h◆Learning Query-Specific Rubrics from Human Preferences for DeepResearch Report Generation8h◆Sim2Signal: Sim-to-Real Benchmarks for Traffic Signal Control8h◆Recursive Value Learning for Long-Horizon Offline Goal-Conditioned RL8h◆Similarity-Aware Personalized Federated Learning in Heterogeneous Environments8h◆Oracle, will I ever learn? A study of prediction convergence and complementarity across link prediction models8h◆UE5M3 FP4 Block Scaling for Stable Language Model Pretraining8h◆Network-Aware Forecasting on Wireless Access Points8h◆Monotonic anomaly detection8h◆Cantelli Constrained Policy Optimization8h◆Secure AI-Driven Super-Resolution for Real-Time Mixed Reality Applications8h◆From Digital to Physical Reservoir Computing: Co-Optimizing Soft Robotic Reservoirs via Dynamics Matching8h◆Optimal Transport for Network Comparison: A Review with Machine Learning Applications8h◆
News/Stride-k Subsampling: Train-Free Audio Token Reduction for Whisper
arxiv
PublishedSeptember 1, 2026 at 4:00 AM

Stride-k Subsampling: Train-Free Audio Token Reduction for Whisper

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

arXiv:2608.30927v1 Announce Type: cross Abstract: Whisper exposes speech through a fixed 1500-token encoder interface, now a default representation for ASR decoders and Whisper-based speech language models (SpeechLMs), yet its redundancy remains largely unexamined. We propose stride-k subsampling, a

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
arxivMeta-ethics and AI: exploring the novel meta-ethical questions in the era of AI8harxivSSAKG 2.0: An Open-Source Package for Structural Associative Sequence Memory and Context-Based Retrieval8harxivEpistemic Sybil Resistance: Multiplying AI Agents Without Multiplying Evidence8harxivDocHop: Benchmarking Out-of-domain Multi-hop Reasoning in Information-Dense Documents8h
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