Zero-Shot RL Just Got Smarter: Adapting Behavioral Foundation Models to Dynamic Environments Behavioral Foundation Models (BFMs) promise zero-shot generalization across tasks—but they stumble when the world changes (e.g., robot joints wear out or environments shift). This
MACHINE LEARNING
-
Model Selection Impact on AI Performance Outcomes
By
–
Which model were you using? It could varies depending on the model
-
Do AI Models Capture Physical Reality Beyond Human Perception?
By
–
would be interesting to discuss! although the mind of course constructs our reality (ala Kant), I think there's no doubt these models are going beyond that and capturing some of the real structure of physical world 'out there' (which our minds have also been evolved to do too)
-
Unlearning or Obfuscating? Benign Relearning of LLMs
By
–
Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning – Machine Learning Blog | ML@CMU | Carnegie Mellon University
-
Snorkel Weekly Demo: Real LLM Workflows Live Q&A
By
–
Seen Snorkel in action yet? Join our weekly demo to explore real LLM workflows plus live Q&A with our ML team. https://
snorkel.ai/webinar/weekly
-demo-with-qa/
…
#GenAI #LLM #AI #SnorkelAI -
Challenges in Multi-Task Learning and Transfer Learning for Atari
By
–
I agree that chasing score on an Atari game has very limited value, but continuous online multi task learning with sparse rewards and transfer learning is definitely not solved for environments even as simple as Atari. Do you not think those are problems worth grappling with?
-
MIT Study Reveals Vision-Language Models Fail with Negation in Medical Imaging
By
–
MIT study reveals that vision-language models, which are often used to analyze medical images to streamline medical diagnosis, can’t handle queries w/negation words like "no" and "not": https://
bit.ly/4dDo9Hj -
Training curve convergence issues in machine learning models
By
–
Watched training curve never converges.
-
Universal Geometry of Embeddings in Deep Learning
By
–
Harnessing the Universal Geometry of Embeddings Rishi Jha, Collin Zhang, Vitaly Shmatikov, John X. Morris: https://
arxiv.org/abs/2505.12540 #ArtificialIntelligence #DeepLearning #MachineLearning -
Universal Geometry of Embeddings in Deep Learning Research
By
–
Harnessing the Universal Geometry of Embeddings Rishi Jha, Collin Zhang, Vitaly Shmatikov, John X. Morris: https://
arxiv.org/abs/2505.12540 #ArtificialIntelligence #DeepLearning #MachineLearning