Tao Gui

AgentGym-RL: An Open-Source Framework to Train LLM Agents for Long-Horizon Decision Making via Multi-Turn RL

R-Horizon: How Far Can Your Large Reasoning Model Really Go in Breadth and Depth?

Game-RL: Synthesizing Multimodal Verifiable Game Data to Boost VLMs' General Reasoning

Critique-RL: Training Language Models For Critiquing Through Two-Stage Reinforcement Learning

BAPO: Stabilizing Off-Policy Reinforcement Learning for LLMs via Balanced Policy Optimization with Adaptive Clipping

Why Reinforcement Fine-Tuning Enables MLLMs Preserve Prior Knowledge Better: A Data Perspective

Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization

Two Minds Better Than One: Collaborative Reward Modeling for LLM Alignment

WorldPM: Understanding Scaling Patterns in Human Preference Modeling

MDAR: A Multi-scene Dynamic Audio Reasoning Benchmark

From Scores to Preferences: Redefining MOS Benchmarking for Speech Quality Reward Modeling

EvaLearn: Quantifying the Learning Capability and Efficiency of LLMs via Sequential Problem Solving

Understanding Parametric and Contextual Knowledge Reconciliation within Large Language Models

Pre-Trained Policy Discriminators are General Reward Models

Have the VLMs Lost Confidence? A Study of Sycophancy in VLMs

Effective Length Extrapolation via Dimension-Wise Positional Embeddings Manipulation

Distill Visual Chart Reasoning Ability from LLMs to MLLMs

RMB: Comprehensively benchmarking reward models in LLM alignment

AgentGym: Evaluating and Evolving Large Language Model-based Agents across Diverse Envronments

INST-IT: Boosting Instance Understanding via Explicit Visual Prompt Instruction Tuning

SPA-VL: A Comprehensive Safety Preference Alignment Dataset for Vision Language Model

Progressive LLM Alignments Using Two-Player Games