paper-with-me

홈 › Papers

EARL: Towards a Unified Analysis-Guided Reinforcement Learning Framework for Egocentric Interaction Reasoning and Pixel Grounding

2026-05-14 · Yuejiao Su, Xinshen Zhang, Zhen Ye, Lei Yao, Lap-Pui Chau, Yi Wang arxiv

Understanding human--environment interactions from egocentric vision is essential for assistive robotics and embodied intelligent agents, yet existing multimodal large language models (MLLMs) still struggle with accurate interaction reasoning and fine-grained pixel grounding. To this end, this paper introduces EARL, an Egocentric Analysis-guided Reinforcement Learning framework that explicitly transfers coarse interaction semantics to query-oriented answering and grounding. Specifically, EARL adopts a two-stage parsing framework including coarse-grained interpretation and fine-grained response. The first stage holistically interprets egocentric interactions and generates a structured textual description. The second stage produces the textual answer and pixel-level mask in response to the user query. To bridge the two stages, we extract a global interaction descriptor as a semantic prior, which is integrated via a novel Analysis-guided Feature Synthesizer (AFS) for query-oriented reasoning. To optimize heterogeneous outputs, including textual answers, bounding boxes, and grounding masks, we design a multi-faceted reward function and train the response stage with GRPO. Experiments on Ego-IRGBench show that EARL achieves 65.48% cIoU for pixel grounding, outperforming previous RL-based methods by 8.37%, while OOD grounding results on EgoHOS indicate strong transferability to unseen egocentric grounding scenarios.

📄 PDF Abstract BibTeX arXiv:2605.14742

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Meta-TTRL: A Metacognitive Framework for Self-Improving Test-Time Reinforcement Learning in Unified Multimodal Models

2026-03-16 · Lit Sin Tan, Junzhe Chen, Xiaolong Fu, Lichen Ma 외 arxiv

Existing test-time scaling (TTS) methods for unified multimodal models (UMMs) in text-to-image (T2I) generation primarily rely on search or sampling strategies that produce only instance-level improvements, limiting the …

Reinforcement Learning

SPARQ: Spiking Early-Exit Neural Networks for Energy-Efficient Edge AI

2026-03-15 · Parth Patne, Mahdi Taheri, Ali Mahani, Maksim Jenihhin 외 arxiv

Spiking neural networks (SNNs) offer inherent energy efficiency due to their event-driven computation model, making them promising for edge AI deployment. However, their practical adoption is limited by the computational…

Reinforcement Learning

A Survey on Rubric-Guided Reinforcement Learning for Language Models

2026-08-27 · Zifei Shan, Fangning Shao arxiv

Reinforcement learning from human feedback (RLHF) has become the dominant paradigm for aligning large language models (LLMs) with human preferences. However, traditional RLHF relies on scalar reward signals that lack int…

Reinforcement Learning

RABot: Reinforcement-Guided Graph Augmentation for Imbalanced and Noisy Social Bot Detection

2026-02-25 · Longlong Zhang, Xi Wang, Haotong Du, Yangyi Xu 외 arxiv

Social bot detection is pivotal for safeguarding the integrity of online information ecosystems. Although recent graph neural network (GNN) solutions achieve strong results, they remain hindered by two practical challeng…

Graph Neural Network

Safety-Guided Flow (SGF): A Unified Framework for Negative Guidance in Safe Generation

2026-03-02 · Mingyu Kim, Young-Heon Kim, Mijung Park arxiv

Safety mechanisms for diffusion and flow models have recently been developed along two distinct paths. In robot planning, control barrier functions are employed to guide generative trajectories away from obstacles at eve…

Image Generation