activity
20242026
collaborators
Showing cs.LGShow all

7 papers · 1 filter

cs.LG2026

Normalized Rewards for Preference Optimization

Shawn Im, Federico Danieli, Skyler Seto +2

Direct Alignment Algorithms (DAAs) such as DPO have become a common way to post-train and align LLMs with human preferences. However, DAAs have been observed to over-optimize their…

cs.LG2026

Cyclical Entropy Eruption: Entropy Dynamics in Agent Reinforcement Learning

Wendi Li, Shawn Im, Sharon Li

Agentic large language models are increasingly used to solve real-world tasks by reasoning over goals, invoking tools, and interacting with external environments. Reinforcement lea…

cs.LG2026

How Well Can Preference Optimization Generalize Under Noisy Feedback?

Shawn Im, Sharon Li

As large language models (LLMs) advance their capabilities, aligning these models with human preferences has become crucial. Preference optimization, which trains models to disting…

cs.LG2026

Towards Interpretability Without Sacrifice: Faithful Dense Layer Decomposition with Mixture of Decoders

James Oldfield, Shawn Im, Sharon Li +3

Multilayer perceptrons (MLPs) are an integral part of large language models, yet their dense representations render them difficult to understand, edit, and steer. Recent methods le…

cs.LG20261 cited

A Unified Understanding and Evaluation of Steering Methods

Shawn Im, Sharon Li

Latent space steering methods provide a practical approach to controlling large language models by applying steering vectors to intermediate activations, guiding outputs toward des…

cs.LG2025

Can DPO Learn Diverse Human Values? A Theoretical Scaling Law

Shawn Im, Sharon Li

Large language models (LLMs) have demonstrated remarkable capabilities but often struggle to align with human preferences, leading to harmful or undesirable outputs. Preference lea…