6 papers
Reflection or Re-Generation? Why LLM Revision Fails Where Human Revision Succeeds
Yefan Tao, Gerald Friedland, Madhusudhanan Chandrasekaran +1
Reflection, the ability to revisit and revise prior reasoning, is central to how humans improve their answers. Large language models (LLMs) are increasingly prompted to "reflect,"…
Self-Aligned Reward: Towards Effective and Efficient Reasoners
Peixuan Han, Adit Krishnan, Gerald Friedland +2
Reinforcement learning with verifiable rewards has significantly advanced reasoning in large language models (LLMs), but such signals remain coarse, offering only binary correctnes…
Declarative Data Pipeline for Large Scale ML Services
Yunzhao Yang, Runhui Wang, Xuanqing Liu +14
Modern distributed data processing systems struggle to balance performance, maintainability, and developer productivity when integrating machine learning at scale. These challenges…
Effects of Feature Correlations on Associative Memory Capacity
Stefan Bielmeier, Gerald Friedland
We investigate how feature correlations influence the capacity of Dense Associative Memory (DAM), a Transformer attention-like model. Practical machine learning scenarios involve f…
Bag of Tricks for Multimodal AutoML with Image, Text, and Tabular Data
Zhiqiang Tang, Zihan Zhong, Tong He +1
This paper studies the best practices for automatic machine learning (AutoML). While previous AutoML efforts have predominantly focused on unimodal data, the multimodal aspect rema…
PPLqa: An Unsupervised Information-Theoretic Quality Metric for Comparing Generative Large Language Models
Gerald Friedland, Xin Huang, Yueying Cui +3
We propose PPLqa, an easy to compute, language independent, information-theoretic metric to measure the quality of responses of generative Large Language Models (LLMs) in an unsupe…