·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
LoongSage: A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated Recipes5h◆SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets5h◆CascadeLedger: Counterfactual Auditing of Information and Resource Integrity in LLM Cascades5h◆N-Gram Memories at Two Bits: Quantizing Lookup Tables in Large Language Models5h◆WEAVE: Weaving Real-World Code Evolution into Dependency-Structured Tasks for Benchmarking and Training Code Agents5h◆Trust but Verify: Designing Region-Inhomogeneous Forward Dynamics for Discrete Diffusion5h◆CGDT-CESN: Correlation-Guided and Dynamics-Trained Clustered Echo State Network for Multivariate Time-Series Prediction5h◆BOLT: Multi-Bottleneck Bound-Guided Learning for Flexible Job-Shop Scheduling with Time Lags and Resource-Coupled Transport5h◆CF-NODE: Counterfactual Spatial Node Intervention for Test-Time Adaptation Under Domain Shift, Evaluated Across Five Diabetic Retinopathy Datasets5h◆Rank Measures of Fine-Tuning Updates Need a Noise Baseline5h◆Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process Rewards5h◆Generative Value Classifier5h◆Predicting Privileged Information Effectiveness for On-Policy Self-Distillation5h◆Don't Retrain, Remix: Adaptation of Web Agents using Mixture Hypernetworks5h◆On the SPOT: Enhancing Speculative Decoding for Diffusion Language Models via Commit-Order Prediction5h◆Deterministic Bounds on Accuracy and Stability of Metric-Space Regression and Classification5h◆Coverage Cliffs in Off-Policy Quantile Certification5h◆Structure-Aware Spatial Kernels for Gaussian Process Distributed Port-Hamiltonian Systems5h◆Initiating Removal from Operational Symptoms in Federated Vehicular Networks5h◆Diagnosing GNN Failure Modes with Layer-Wise Probe Entropy5h◆LoongSage: A Production-Grade, High-Efficiency Agentic RL Framework for Frontier LLMs with Validated Recipes5h◆SEA-HELM Audio: Evaluating Multilingual ASR-LLM Pipeline Performance Across SEA-HELM Subsets5h◆CascadeLedger: Counterfactual Auditing of Information and Resource Integrity in LLM Cascades5h◆N-Gram Memories at Two Bits: Quantizing Lookup Tables in Large Language Models5h◆WEAVE: Weaving Real-World Code Evolution into Dependency-Structured Tasks for Benchmarking and Training Code Agents5h◆Trust but Verify: Designing Region-Inhomogeneous Forward Dynamics for Discrete Diffusion5h◆CGDT-CESN: Correlation-Guided and Dynamics-Trained Clustered Echo State Network for Multivariate Time-Series Prediction5h◆BOLT: Multi-Bottleneck Bound-Guided Learning for Flexible Job-Shop Scheduling with Time Lags and Resource-Coupled Transport5h◆CF-NODE: Counterfactual Spatial Node Intervention for Test-Time Adaptation Under Domain Shift, Evaluated Across Five Diabetic Retinopathy Datasets5h◆Rank Measures of Fine-Tuning Updates Need a Noise Baseline5h◆Lean4AgenticRL: Advancing Agentic RL with Verifiable Multi-level Process Rewards5h◆Generative Value Classifier5h◆Predicting Privileged Information Effectiveness for On-Policy Self-Distillation5h◆Don't Retrain, Remix: Adaptation of Web Agents using Mixture Hypernetworks5h◆On the SPOT: Enhancing Speculative Decoding for Diffusion Language Models via Commit-Order Prediction5h◆Deterministic Bounds on Accuracy and Stability of Metric-Space Regression and Classification5h◆Coverage Cliffs in Off-Policy Quantile Certification5h◆Structure-Aware Spatial Kernels for Gaussian Process Distributed Port-Hamiltonian Systems5h◆Initiating Removal from Operational Symptoms in Federated Vehicular Networks5h◆Diagnosing GNN Failure Modes with Layer-Wise Probe Entropy5h◆
News/CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition
arxiv
PublishedOctober 15, 2026 at 4:00 AM

CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition

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

arXiv:2610.19134v1 Announce Type: new Abstract: Feature distillation is usually expected to add to logit distillation. We find that this expectation does not hold for two common feature losses in compact multilingual automatic speech recognition (ASR), and analyze why through their gradients. The gr

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
arxivRCPN: towards robust pseudo-labels in unsupervised CLIP adaptation via consistent prototypes and neighborhoods5harxivCan Graph Learning Learn Circuits?5harxivBeyond Fixed Acquisition Functions: Temporal Difference Active Learning for Perturbation Discovery5harxivFrom Events to Reasoning: Hierarchical Rule Composition over Explicit States5h
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