5 papers
Inference-Time Distillation: Cost-Efficient Agents Without Fine-Tuning or Manual Prompt Engineering
Vishnu Sarukkai, Asanshay Gupta, James Hong +2
Deploying LLM agents at scale typically requires choosing between quality and cost. Existing cost-reduction approaches fail to preserve agility: the ability to iterate rapidly with…
Learning to Ball: Composing Policies for Long-Horizon Basketball Moves
Pei Xu, Zhen Wu, Ruocheng Wang +5
Learning a control policy for a multi-phase, long-horizon task, such as basketball maneuvers, remains challenging for reinforcement learning approaches due to the need for seamless…
Self-Generated In-Context Examples Improve LLM Agents for Sequential Decision-Making Tasks
Vishnu Sarukkai, Zhiqiang Xie, Kayvon Fatahalian
Improving Large Language Model (LLM) agents for sequential decision-making tasks typically requires extensive task-specific knowledge engineering--custom prompts, curated examples,…
Automated Rewards via LLM-Generated Progress Functions
Vishnu Sarukkai, Brennan Shacklett, Zander Majercik +3
Large Language Models (LLMs) have the potential to automate reward engineering by leveraging their broad domain knowledge across various tasks. However, they often need many iterat…
Block and Detail: Scaffolding Sketch-to-Image Generation
Vishnu Sarukkai, Lu Yuan, Mia Tang +2
We introduce a novel sketch-to-image tool that aligns with the iterative refinement process of artists. Our tool lets users sketch blocking strokes to coarsely represent the placem…