·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Anthropic apologizes for invisible Claude Fable guardrails31m◆Google DeepMind is worried about what happens when millions of agents start to interact1h◆Deezer launches an AI music detector for other streaming services4h◆Opendoor’s India exit is fueling a bigger conversation about AI and outsourcing8h◆MODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning8h◆Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!8h◆Generalizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Solutions8h◆The Impossibility of Eliciting Latent Knowledge8h◆Mapping Scientific Literature with Large Language Models and Topic Modeling8h◆Grounding Computer Use Agents on Human Demonstrations8h◆Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models8h◆LSTM based IoT Device Identification8h◆StanceNakba Shared Task: Actor and Topic-Aware Stance Detection in Public Discourse8h◆Breaking the Ice: Analyzing Cold Start Latency in vLLM8h◆Modeling Complex Behaviors: Multi-Personality Composition and Dynamic Switching in Vision-Language Models8h◆DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?8h◆Higher order PCA-like rotation-invariant features for detailed shape descriptors modulo rotation8h◆CoVEBench: Can Video Editing Models Handle Complex Instructions?8h◆Minimal surfaces, Knots, and Neural Networks8h◆CCKS: Consensus-based Communication and Knowledge Sharing8h◆Anthropic apologizes for invisible Claude Fable guardrails31m◆Google DeepMind is worried about what happens when millions of agents start to interact1h◆Deezer launches an AI music detector for other streaming services4h◆Opendoor’s India exit is fueling a bigger conversation about AI and outsourcing8h◆MODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning8h◆Position: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!8h◆Generalizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Solutions8h◆The Impossibility of Eliciting Latent Knowledge8h◆Mapping Scientific Literature with Large Language Models and Topic Modeling8h◆Grounding Computer Use Agents on Human Demonstrations8h◆Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models8h◆LSTM based IoT Device Identification8h◆StanceNakba Shared Task: Actor and Topic-Aware Stance Detection in Public Discourse8h◆Breaking the Ice: Analyzing Cold Start Latency in vLLM8h◆Modeling Complex Behaviors: Multi-Personality Composition and Dynamic Switching in Vision-Language Models8h◆DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?8h◆Higher order PCA-like rotation-invariant features for detailed shape descriptors modulo rotation8h◆CoVEBench: Can Video Editing Models Handle Complex Instructions?8h◆Minimal surfaces, Knots, and Neural Networks8h◆CCKS: Consensus-based Communication and Knowledge Sharing8h◆
News/Routing-Aware Expert Calibration for Machine Unlearning in Mixture-of-Experts Language Models
arxiv
PublishedJune 10, 2026 at 4:00 AM
—neutral

Routing-Aware Expert Calibration for Machine Unlearning in Mixture-of-Experts Language Models

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

arXiv:2606.10338v1 Announce Type: cross Abstract: Machine unlearning is increasingly important for large language models, yet unlearning in Mixture-of-Experts (MoE) architectures remains underexplored. Unlike dense models, MoE architectures employ a router at each layer to assign each token to a spa

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
arxivMODF-SIR: A Multi-agent Omni-modal Distilled Framework for Social Intelligence Reasoning8harxivPosition: Stop Anthropomorphizing Intermediate Tokens as Reasoning/Thinking Traces!8harxivGeneralizing Beyond Suboptimality: Offline Reinforcement Learning Learns Effective Scheduling through Random Solutions8harxivThe Impossibility of Eliciting Latent Knowledge8h
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