collaborators

6 papers

cs.LG2025

RL's Razor: Why Online Reinforcement Learning Forgets Less

Idan Shenfeld, Jyothish Pari, Pulkit Agrawal

Comparison of fine-tuning models with reinforcement learning (RL) and supervised fine-tuning (SFT) reveals that, despite similar performance at a new task, RL preserves prior knowl…

stat.ML2025

Best-of-N through the Smoothing Lens: KL Divergence and Regret Analysis

Gholamali Aminian, Idan Shenfeld, Amir R. Asadi +2

A simple yet effective method for inference-time alignment of generative models is Best-of- (BoN), where outcomes are sampled from a reference policy, evaluated using a prox…

cs.LG2025

Language Model Personalization via Reward Factorization

Idan Shenfeld, Felix Faltings, Pulkit Agrawal +1

Modern large language models (LLMs) are optimized for human-aligned responses using Reinforcement Learning from Human Feedback (RLHF). However, existing RLHF approaches assume a un…

cs.LG2025

KL-Regularized RLHF with Multiple Reference Models: Exact Solutions and Sample Complexity

Gholamali Aminian, Amir R. Asadi, Idan Shenfeld +1

Recent methods for aligning large language models (LLMs) with human feedback predominantly rely on a single reference model, which limits diversity, model overfitting, and underuti…

cs.LG2024

Learning How Hard to Think: Input-Adaptive Allocation of LM Computation

Mehul Damani, Idan Shenfeld, Andi Peng +2

Computationally intensive decoding procedures--including search, reranking, and self-critique--can improve the quality of language model (LM) outputs in problems spanning code gene…

cs.HC2024

The Future of Open Human Feedback

Shachar Don-Yehiya, Ben Burtenshaw, Ramon Fernandez Astudillo +17

Human feedback on conversations with language language models (LLMs) is central to how these systems learn about the world, improve their capabilities, and are steered toward desir…