Work

Imbalanced Gradients in RL Post-Training of Multi-Task LLMs

Year: 2026

Type: conference-paper

Abstract: Runzhe Wu, Ankur Samanta, Ayush Jain, Scott Fujimoto, Jeongyeol Kwon, Ben Kretzu, Youliang Yu, Kaveh Hassani, Boris Vidolov, Yonathan Efroni. Findings of the Association for Computational Linguistics: EACL 2026. 2026.

Source: Findings of the Association for Computational Linguistics: EACL

DOI: 10.18653/v1/2026.findings-eacl.164

Authors: Runzhe Wu, Ankur Samanta, Ayush Jain, Scott Fujimoto, Jeongyeol Kwon, Ben Kretzu, Youliang Yu, Kaveh Hassani, Boris Vidolov, Yonathan Efroni

Institutions: Cornell University, Columbia University, Technion – Israel Institute of Technology, Tel Aviv University

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

OpenAlex ID: W7140078914

FWCI:

Cites:

Cited by:

Related to: 10