·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance2h◆Pixel-Level Transformers in Remote Sensing: A Canopy Height Case Study2h◆Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents2h◆Boosting Adversarial Robustness and Generalization with Dictionary Structure2h◆Making Duplicate Reimbursement Unrepresentable: A Verified Ethereum E-Invoice System for Humans and AI Agents2h◆HandAnthro: Automated Hand Anthropometry from a Single Image2h◆AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems2h◆A Dominant Supplier Slows Recursive Drift More Than It Steers It2h◆COMiT: Learning Structured Visual Tokens through Sequential Communication2h◆PAC-CF: Calibrating Irreversible Frontier Pruning in LLM-Guided Search2h◆RobotValues: Evaluating Household Robots When Human Values Conflict2h◆TacForcing: Streaming Action Generation with Execution-Time Tactile Feedback2h◆Learning Spectrally Optimised Mesh-Free Discretisations2h◆Question-Specific Knowledge Graphs for Efficient Visual Reasoning2h◆Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos2h◆Local Search with Correlated Randomness2h◆When Can Prefixes Compile LoRA? Exact Resource-Capped Tests for Frozen Attention2h◆How Can Recommendation Feedback Evolve Agent Memory?2h◆Sage: Formalization with Semantic Correction2h◆Alignment Forecasting: Predicting Misalignment From Training Data2h◆Predictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance2h◆Pixel-Level Transformers in Remote Sensing: A Canopy Height Case Study2h◆Explore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents2h◆Boosting Adversarial Robustness and Generalization with Dictionary Structure2h◆Making Duplicate Reimbursement Unrepresentable: A Verified Ethereum E-Invoice System for Humans and AI Agents2h◆HandAnthro: Automated Hand Anthropometry from a Single Image2h◆AgentBug-Smith: Automatically Reproducing Real-World Harness Bugs in Agentic Systems2h◆A Dominant Supplier Slows Recursive Drift More Than It Steers It2h◆COMiT: Learning Structured Visual Tokens through Sequential Communication2h◆PAC-CF: Calibrating Irreversible Frontier Pruning in LLM-Guided Search2h◆RobotValues: Evaluating Household Robots When Human Values Conflict2h◆TacForcing: Streaming Action Generation with Execution-Time Tactile Feedback2h◆Learning Spectrally Optimised Mesh-Free Discretisations2h◆Question-Specific Knowledge Graphs for Efficient Visual Reasoning2h◆Dropout Universality: Scaling Laws and Optimal Scheduling at the Edge-of-Chaos2h◆Local Search with Correlated Randomness2h◆When Can Prefixes Compile LoRA? Exact Resource-Capped Tests for Frozen Attention2h◆How Can Recommendation Feedback Evolve Agent Memory?2h◆Sage: Formalization with Semantic Correction2h◆Alignment Forecasting: Predicting Misalignment From Training Data2h◆
News/$S^3$: Spectral Null-Space Swap Makes Reasoning Models Efficient
arxiv
PublishedSeptember 30, 2026 at 4:00 AM

$S^3$: Spectral Null-Space Swap Makes Reasoning Models Efficient

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

arXiv:2609.37976v1 Announce Type: cross Abstract: LLMs trained with Chain-of-thought excel in reasoning capability, but often come with excessive token cost. We find that the core of reasoning capacity lies in the Thinking model's weight component within the null space of a projection defined by the

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
arxivPredictive Self-Supervised Learning Provably Identifies Stochastic Signals under Nuisance2harxivPixel-Level Transformers in Remote Sensing: A Canopy Height Case Study2harxivExplore, Execute, Evolve: A Skill Acquisition and Reuse Loop for Embodied Agents2harxivBoosting Adversarial Robustness and Generalization with Dictionary Structure2h
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 ↗
Built by Marouane Gazouzi
HomeModelsNews