·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Perplexity trusts GPT-6 Astra with end-to-end systems-2397m◆TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents4h◆Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation4h◆SPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models4h◆Towards a Deterministic Math Solver for Clinical Language Models4h◆Demystifying the Privacy-Utility Trade-off in LLM Interactions4h◆Amulet: a Python Library for Assessing Interactions Among ML Defenses and Risks4h◆Automating Quadratic Unconstrained Binary Optimization (QUBO) Formulation Generation from Natural Language4h◆Defining AI Agents: A Compendium of Criteria, Metrics, and Benchmarks4h◆Finishing the Task Is Not Enough: Evaluating Agent Resilience and Considerate Participation under Accumulating Challenge4h◆The Agent Incident Registry: Toward Preventing Repeated AI Agent Failures4h◆Decoupling Readiness from Release for Tail-Aware Scheduling of Agentic LLM Workflows4h◆Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents4h◆When Validation Stops Learning: Auditing Update Admission for Continual Embodied Agents4h◆DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat4h◆Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning4h◆From Queries to Narratives: Cultural Heritage Data Stories for Knowledge Graph Exploration and Quality Assessment4h◆LLMs as Post-hoc Auditors of Physiological Plausibility in Symbolic Regression: A Clinician-Evaluated Case Study4h◆Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration4h◆RouteRepair: Instance-Level Failure Diagnosis and Targeted Repair in LLM-Based Automated Heuristic Design for Routing Optimization4h◆Perplexity trusts GPT-6 Astra with end-to-end systems-2397m◆TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents4h◆Natural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation4h◆SPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models4h◆Towards a Deterministic Math Solver for Clinical Language Models4h◆Demystifying the Privacy-Utility Trade-off in LLM Interactions4h◆Amulet: a Python Library for Assessing Interactions Among ML Defenses and Risks4h◆Automating Quadratic Unconstrained Binary Optimization (QUBO) Formulation Generation from Natural Language4h◆Defining AI Agents: A Compendium of Criteria, Metrics, and Benchmarks4h◆Finishing the Task Is Not Enough: Evaluating Agent Resilience and Considerate Participation under Accumulating Challenge4h◆The Agent Incident Registry: Toward Preventing Repeated AI Agent Failures4h◆Decoupling Readiness from Release for Tail-Aware Scheduling of Agentic LLM Workflows4h◆Grounding Agent Memory: Environment-Probing Curation for Enterprise Agents4h◆When Validation Stops Learning: Auditing Update Admission for Continual Embodied Agents4h◆DRG-MAPPO: Hierarchical Dynamic Role-Graph Multi-Agent Reinforcement Learning for Cooperative Air Combat4h◆Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning4h◆From Queries to Narratives: Cultural Heritage Data Stories for Knowledge Graph Exploration and Quality Assessment4h◆LLMs as Post-hoc Auditors of Physiological Plausibility in Symbolic Regression: A Clinician-Evaluated Case Study4h◆Calibration-Aware Uncertainty Cascades for Efficient Heterogeneous Model Collaboration4h◆RouteRepair: Instance-Level Failure Diagnosis and Targeted Repair in LLM-Based Automated Heuristic Design for Routing Optimization4h◆
News/Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning
arxiv
PublishedSeptember 12, 2026 at 4:00 AM
—neutral

Beyond Confidence: Stability-Aware Test-Time Adaptation for LLM Reasoning

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

arXiv:2609.11393v1 Announce Type: new Abstract: Test-time adaptation has emerged as a lightweight alternative to costly post-training for improving the reasoning capabilities of Large Language Models (LLMs) on downstream tasks. Predictive entropy provides a model-derived signal for such adaptation,

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
arxivTeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents4harxivNatural Language Access to Domain-Specific Metadata: A Reusable Framework for LLM Query Generation4harxivSPECTRA: Band-Routed Embedding and Stage-Wise LoRA for Cross-Sensor Fine-Tuning of Geospatial Foundation Models4harxivTowards a Deterministic Math Solver for Clinical Language Models4h
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