6 papers
SPIRAL: Learning to Search and Aggregate
Jubayer Ibn Hamid, Ifdita Hasan Orney, Michael Y. Li +5
Language model reasoning can be substantially improved at test time via scaffolds that scale inference compute across different primitives -- sequential reasoning within a trace, i…
Test-Time Alignment via Hypothesis Reweighting
Yoonho Lee, Jonathan Williams, Henrik Marklund +4
Reward models trained on aggregate preferences often fail to capture individual users' values, but existing adaptation methods such as fine-tuning or long-context conditioning are…
Meta-Harness: End-to-End Optimization of Model Harnesses
Yoonho Lee, Roshen Nair, Qizheng Zhang +3
The performance of large language model (LLM) systems depends not only on model weights, but also on their harness: the code that determines what information to store, retrieve, an…
Feedback Descent: Open-Ended Text Optimization via Pairwise Comparison
Yoonho Lee, Joseph Boen, Chelsea Finn
We introduce \textit{Feedback Descent}, a framework that optimizes text artifacts -- prompts, code, and molecules -- through structured textual feedback, rather than relying solely…
RLAD: Training LLMs to Discover Abstractions for Solving Reasoning Problems
Yuxiao Qu, Anikait Singh, Yoonho Lee +4
Reasoning requires going beyond pattern matching or memorization of solutions to identify and implement "algorithmic procedures" that can be used to deduce answers to hard problems…
Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling
Yuejiang Liu, Jubayer Ibn Hamid, Annie Xie +3
Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its…