·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Replit’s CEO says the future of work is the ‘self-driving company’56m◆Bridging the Gap: Converting Read Text to Conversational Dialogue9h◆Sharp and Stable Label-Free Calibration Certificates from Logical Coherence9h◆Backward Compatibility in Zero-Bit Neural Audio Watermarking9h◆Disagreement-Guided Weak-to-Strong Routing9h◆One Is Not Enough: How People Use Multiple AI Models in Everyday Life9h◆L2G-SR: Learning to Guide LLM-based Symbolic Regression9h◆Kolab: Benchmarking Human-Agent Collaboration in Long-Horizon Workflows9h◆CompMat-Bench: Benchmarking AI Agents for Computational Materials Science9h◆Trajectory-Level Speculation with Semantic Verification for Tool-Using Agents9h◆Unfragmenting Nawatl NMT: Single-Variant Specialization for Masewalkopa9h◆LLMR: Knowledge Distillation with a Large Language Model-Induced Reward9h◆PRISM: PRIvileged Sibling-guided Multi-Agent Topology Design9h◆Training Reasoning Models to Generate Diverse Scientific Ideas9h◆Linear Independence of Polynomial Compositions with Bounded Input Degree9h◆Do LLMs have core beliefs?9h◆KnowGPO: Knowledge-Grouped Policy Optimization for Search Agents9h◆Boltzmann generators for amorphous particle systems9h◆Diffusion Models in Simulation-Based Inference: A Tutorial Review9h◆Sim-to-Real Transfer for Muscle-Actuated Robots via Generalized Actuator Networks9h◆Replit’s CEO says the future of work is the ‘self-driving company’56m◆Bridging the Gap: Converting Read Text to Conversational Dialogue9h◆Sharp and Stable Label-Free Calibration Certificates from Logical Coherence9h◆Backward Compatibility in Zero-Bit Neural Audio Watermarking9h◆Disagreement-Guided Weak-to-Strong Routing9h◆One Is Not Enough: How People Use Multiple AI Models in Everyday Life9h◆L2G-SR: Learning to Guide LLM-based Symbolic Regression9h◆Kolab: Benchmarking Human-Agent Collaboration in Long-Horizon Workflows9h◆CompMat-Bench: Benchmarking AI Agents for Computational Materials Science9h◆Trajectory-Level Speculation with Semantic Verification for Tool-Using Agents9h◆Unfragmenting Nawatl NMT: Single-Variant Specialization for Masewalkopa9h◆LLMR: Knowledge Distillation with a Large Language Model-Induced Reward9h◆PRISM: PRIvileged Sibling-guided Multi-Agent Topology Design9h◆Training Reasoning Models to Generate Diverse Scientific Ideas9h◆Linear Independence of Polynomial Compositions with Bounded Input Degree9h◆Do LLMs have core beliefs?9h◆KnowGPO: Knowledge-Grouped Policy Optimization for Search Agents9h◆Boltzmann generators for amorphous particle systems9h◆Diffusion Models in Simulation-Based Inference: A Tutorial Review9h◆Sim-to-Real Transfer for Muscle-Actuated Robots via Generalized Actuator Networks9h◆
News/Pareto-Optimal Offline Reinforcement Learning via Smooth Tchebycheff Scalarization
arxiv
PublishedSeptember 15, 2026 at 4:00 AM

Pareto-Optimal Offline Reinforcement Learning via Smooth Tchebycheff Scalarization

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

arXiv:2604.13175v2 Announce Type: replace-cross Abstract: Large language models can be aligned with human preferences through offline reinforcement learning (RL) on small labeled datasets. While single-objective alignment is well-studied, many real-world applications demand the simultaneous optimiza

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
arxivBridging the Gap: Converting Read Text to Conversational Dialogue9harxivSharp and Stable Label-Free Calibration Certificates from Logical Coherence9harxivBackward Compatibility in Zero-Bit Neural Audio Watermarking9harxivDisagreement-Guided Weak-to-Strong Routing9h
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