#Model Training
Latest news articles tagged with "Model Training". Follow the timeline of events, related topics, and entities.
Articles (5)
-
🇺🇸 Specification-Aware Distribution Shaping for Robotics Foundation Models
[USA]
arXiv:2603.17969v1 Announce Type: cross Abstract: Robotics foundation models have demonstrated strong capabilities in executing natural language instructions across diverse tasks and environments. Ho...
Related: #Robotics AI -
🇺🇸 Alignment Makes Language Models Normative, Not Descriptive
[USA]
arXiv:2603.17218v1 Announce Type: cross Abstract: Post-training alignment optimizes language models to match human preference signals, but this objective is not equivalent to modeling observed human ...
Related: #AI Ethics -
🇺🇸 Compression Favors Consistency, Not Truth: When and Why Language Models Prefer Correct Information
[USA]
arXiv:2603.11749v1 Announce Type: cross Abstract: Why do language models sometimes prefer correct statements even when trained on mixed-quality data? We introduce the Compression--Consistency Princip...
Related: #AI Limitations -
🇺🇸 A Quantitative Characterization of Forgetting in Post-Training
[USA]
arXiv:2603.12163v1 Announce Type: cross Abstract: Continual post-training of generative models is widely used, yet a principled understanding of when and why forgetting occurs remains limited. We dev...
Related: #Machine Learning -
🇺🇸 Post-Training with Policy Gradients: Optimality and the Base Model Barrier
[USA]
arXiv:2603.06957v1 Announce Type: cross Abstract: We study post-training linear autoregressive models with outcome and process rewards. Given a context $\boldsymbol{x}$, the model must predict the re...
Related: #AI Optimization
Key Entities (2)
- Algorithmic bias (1 news)
- AI safety (1 news)
About the topic: Model Training
The topic "Model Training" aggregates 5+ news articles from various countries.