This repository contains the Qwen3-VL-4B weights fine-tuned with GRPO, released as a baseline for the paper Learning Spatiotemporal Sensitivity in Video LLMs via Counterfactual Reinforcement Learning.

This checkpoint is a Group Relative Policy Optimization (GRPO) baseline fine-tuned on top of Qwen3-VL for video question answering, released as a reference RL baseline for the CRPO paper.

Video large language models (Video LLMs) often answer video questions through shortcuts such as single-frame cues and language priors rather than by tracking spatiotemporal dynamics. The CRPO paper studies this problem and compares against multiple RL baselines, of which this checkpoint is one.

The model is evaluated on DyBench, a paired counterfactual video benchmark with 3,014 videos covering reversible dynamics, moving direction, and event sequence, together with standard video-QA benchmarks (Video-MME, TempCompass, MVBench, TimeBlind). See the paper for full numbers and comparison with CRPO.