Moksh Jain

PhD student@Université de Montréal·OpenReview

研究方向

Probabilistic Inference · Bayesian Deep Learning · Experimental Design · Reinforcement Learning · Deep Learning · Anomaly Detection · Model Compression

5.5

Trajectory Balance with Asynchrony: Decoupling Exploration and Learning for Fast, Scalable LLM Post-Training

NeurIPS 2025Poster

7.0

Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning

ICLR 2025Poster

通讯

6.8

Action abstractions for amortized sampling

ICLR 2025Poster

4.8

Automated Discovery of Pairwise Interactions from Unstructured Data

合作者 (20)

Moksh Jain

Latent Veracity Inference for Identifying Errors in Stepwise Reasoning

A Comedy of Estimators: On KL Regularization in RL Training of LLMs

Benchmarking World-Model Learning

Recursive Self-Aggregation Unlocks Deep Thinking in Large Language Models

Bayesian Symbolic Regression with Entropic Reinforcement Learning

Trajectory Balance with Asynchrony: Decoupling Exploration and Learning for Fast, Scalable LLM Post-Training

Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning

Action abstractions for amortized sampling

Automated Discovery of Pairwise Interactions from Unstructured Data