9 papers · 1 filter
RRPO: Reference-Relative Policy Optimization with Stratified Conditional Rollouts
Yuxin Xiong, Xunyi Jiang, Rohan Surana +8
Group Relative Policy Optimization (GRPO) has shown strong effectiveness in reinforcement learning from verifiable feedback, where sampled rollouts can be compared within a group u…
F-GRPO: Factorized Group-Relative Policy Optimization for Unified Candidate Generation and Ranking
Rohan Surana, Gagan Mundada, Junda Wu +9
Traditional retrieval pipelines optimize utility through stages of candidate retrieval and reranking, where ranking operates over a predefined candidate set. Large Language Models…
MASS-DPO: Multi-negative Active Sample Selection for Direct Policy Optimization
Rohan Surana, Xintong Li, Sheldon Yu +7
Multi-negative preference optimization under the Plackett--Luce (PL) model extends Direct Preference Optimization (DPO) by leveraging comparative signals across one preferred and m…
Skill-R1: Agent Skill Evolution via Reinforcement Learning
Yash Vishe, Rohan Surana, Xunyi Jiang +8
Agentic large language models often rely on skills, reusable natural language procedures that guide planning, action, and tool use. In practice, skills are typically improved throu…
Skill-CMIB: Multimodal Agent Skill for Consistent Action via Conditional Multimodal Information Bottleneck
Zihan Huang, Junda Wu, Tong Yu +6
While LLM-based agents excel at planning and executing long action sequences, their execution often remains inconsistent across trials, limiting reliability. Consolidating agent co…
Generate, Filter, Control, Replay: A Comprehensive Survey of Rollout Strategies for LLM Reinforcement Learning
Rohan Surana, Gagan Mundada, Xunyi Jiang +19
Reinforcement learning (RL) has become a central post-training tool for improving the reasoning abilities of large language models (LLMs). In these systems, the rollout, the trajec…
WS-GRPO: Weakly-Supervised Group-Relative Policy Optimization for Rollout-Efficient Reasoning
Gagan Mundada, Zihan Huang, Rohan Surana +8
Group Relative Policy Optimization (GRPO) is effective for training language models on complex reasoning. However, since the objective is defined relative to a group of sampled tra…
AMPS: Adaptive Modality Preference Steering via Functional Entropy
Zihan Huang, Xintong Li, Rohan Surana +5
Multimodal Large Language Models (MLLMs) often exhibit significant modality preference, which is a tendency to favor one modality over another. Depending on the input, they may ove…
In-context Ranking Preference Optimization
Junda Wu, Rohan Surana, Zhouhang Xie +6
Recent developments in Direct Preference Optimization (DPO) allow large language models (LLMs) to function as implicit ranking models by maximizing the margin between preferred and…