Improving Thinking Process in Visual Grounding via Free Thinking Rewards
Lists
Only you see this
Abstract
We study egocentric visual intention grounding, where an assistant must infer and localize the object implied by a first-person view and an intention sentence without explicit object naming. Existing approaches either use a two-stage reasoning–then-grounding pipeline or apply reinforcement learning (RL) to train think-then-answer VLMs, but they optimize only IoU-based rewards, which often leads to reward hacking, improving box accuracy while neglecting reasoning quality. We introduce a label-free thinking-process reward that needs no human Chain-of-Thought labels or teacher models; it scores each sampled reasoning trace by how much it increases the likelihood of the correct answer, thus favoring reasoning that truly supports prediction. We also propose a data filtering strategy that selects informative easy-to-medium samples for RL using rollout error rate and reward variance. Together, these form a general recipe for process-aware RL finetuning of vision-language assistants for egocentric intention grounding. Our method achieves new state-of-the-art results on EgoIntention, boosting [email protected] by +3.2 (3B) and +2.1 (7B) over strong Qwen2.5-VL baselines, and generalizes zero-shot to RefEgo-Int with +10.2 (3B) and +7.5 (7B).
BibTeX
@misc{akula2026improving,
author = {Arjun Reddy Akula},
title = {{Improving Thinking Process in Visual Grounding via Free Thinking Rewards}},
year = {2026},
url = {https://openreview.net/forum?id=YsdYTxkw27},
howpublished = {OpenReview},
}Loading related content...