5 papers
Posterior Refinement: Fast Language Generation via Any-Order Flow Maps
Manan Agarwal, Sheel Shah, Chanhyuk Lee +6
Non-autoregressive generation offers a powerful paradigm for iterative refinement, allowing models to recursively critique, erase and regenerate arbitrary subsets of tokens. Howeve…
How to Guide Your Flow: Few-Step Alignment via Flow Map Reward Guidance
Jerry Y. Huang, Justin Lin, Sheel Shah +2
In generative modeling, we often wish to produce samples that maximize a user-specified reward such as aesthetic quality or alignment with human preferences, a problem known as \te…
Flow Map Language Models: One-step Language Modeling via Continuous Denoising
Chanhyuk Lee, Jaehoon Yoo, Manan Agarwal +6
Language models based on discrete diffusion have attracted widespread interest for their potential to provide faster generation than autoregressive models. Despite their promise, t…
A View of the Certainty-Equivalence Method for PAC RL as an Application of the Trajectory Tree Method
Shivaram Kalyanakrishnan, Sheel Shah, Santhosh Kumar Guguloth
Reinforcement learning (RL) enables an agent interacting with an unknown MDP to optimise its behaviour by observing transitions sampled from . A natural entity that emerges…
On the Regret of Online Coded Caching
Anupam Nayak, Sheel Shah, Nikhil Karamchandani
We consider the widely studied problem of coded caching under non-uniform requests where users independently request files according to some underlying popularity distribution in e…