·
DataBubble
  • Home
  • Models
  • News
  • Compare
  • Boards
  • Pricing
  • About
  • Newsletter
  • Methodology
  • Contact
Latest
Right Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT59m◆SAGE: A Statistical Acceptance Gate for Self-Evolving Agents59m◆PowerZooJax: A JAX-based Power System Benchmark for Reinforcement Learning59m◆GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning59m◆Mirror-Score: Calibrated, Inference-only Scoring Exposes the Limits of Sequence-compatibility Ranking in D-peptide Design59m◆SMat-Attention: Structured Long-Context Sequence Modeling59m◆LongCat-DeepResearch Technical Report59m◆A Polyphonic Conception of AI Understanding59m◆GeoOutageBench: Benchmarking Ambiguity-aware, Ontology-grounded Geospatiotemporal KGQA for Multimodal Power Outage and Resilience Analysis59m◆An Exact Generate - Transform Decomposition of Small-LLM Team Scaling Across Orchestration Architectures59m◆The Layer Mystery of VLA: An Information-Theoretical Analysis of VLA Latent Interface59m◆AdaST: Adaptive Coupling for Spatial-Temporal Forecasting59m◆Memory Is a Derivation: The Distributed-Evidence Paradox in Long-Term Agents59m◆More Features Are Not More Evidence: Limits of Training-Free Human Activity Recognition with Jev59m◆Principled Thoughts for Latent Recursive LLM Systems59m◆FigAct: Turning Scientific Figures into Active Canvases for Explanation59m◆An Empirical Study and Assessment of EU AI Act Compliance Checkers59m◆MERID: Multimodal Exploration via Recursive Self-Improvement Agents for Major Depression Analysis59m◆ChronoSRL: Temporal Geometry for Self-Supervised Reinforcement Learning59m◆CoRe: Co-Evolving Reward Models for Mitigating Latent Reward Hacking in Video Diffusion Models59m◆Right Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT59m◆SAGE: A Statistical Acceptance Gate for Self-Evolving Agents59m◆PowerZooJax: A JAX-based Power System Benchmark for Reinforcement Learning59m◆GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning59m◆Mirror-Score: Calibrated, Inference-only Scoring Exposes the Limits of Sequence-compatibility Ranking in D-peptide Design59m◆SMat-Attention: Structured Long-Context Sequence Modeling59m◆LongCat-DeepResearch Technical Report59m◆A Polyphonic Conception of AI Understanding59m◆GeoOutageBench: Benchmarking Ambiguity-aware, Ontology-grounded Geospatiotemporal KGQA for Multimodal Power Outage and Resilience Analysis59m◆An Exact Generate - Transform Decomposition of Small-LLM Team Scaling Across Orchestration Architectures59m◆The Layer Mystery of VLA: An Information-Theoretical Analysis of VLA Latent Interface59m◆AdaST: Adaptive Coupling for Spatial-Temporal Forecasting59m◆Memory Is a Derivation: The Distributed-Evidence Paradox in Long-Term Agents59m◆More Features Are Not More Evidence: Limits of Training-Free Human Activity Recognition with Jev59m◆Principled Thoughts for Latent Recursive LLM Systems59m◆FigAct: Turning Scientific Figures into Active Canvases for Explanation59m◆An Empirical Study and Assessment of EU AI Act Compliance Checkers59m◆MERID: Multimodal Exploration via Recursive Self-Improvement Agents for Major Depression Analysis59m◆ChronoSRL: Temporal Geometry for Self-Supervised Reinforcement Learning59m◆CoRe: Co-Evolving Reward Models for Mitigating Latent Reward Hacking in Video Diffusion Models59m◆
News/CoRe: Co-Evolving Reward Models for Mitigating Latent Reward Hacking in Video Diffusion Models
arxiv
PublishedSeptember 30, 2026 at 4:00 AM

CoRe: Co-Evolving Reward Models for Mitigating Latent Reward Hacking in Video Diffusion Models

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

arXiv:2609.36245v1 Announce Type: new Abstract: Latent reward models (LRMs) enable efficient alignment of video diffusion models by scoring intermediate states directly in latent space. However, we find that optimizing against a fixed latent reward rapidly leads to latent reward hacking: the predict

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
arxivRight Words, Wrong Moment: A Clinician-Grounded Analysis of Distress in 19,930 Conversations between Young People and ChatGPT59marxivSAGE: A Statistical Acceptance Gate for Self-Evolving Agents59marxivPowerZooJax: A JAX-based Power System Benchmark for Reinforcement Learning59marxivGeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning59m
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