pith. sign in

arxiv: 2405.10292 · v3 · pith:EY3P7OOOnew · submitted 2024-05-16 · 💻 cs.AI · cs.CL· cs.CV· cs.LG

Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning

classification 💻 cs.AI cs.CLcs.CVcs.LG
keywords reasoningframeworkmodelsagentsdecision-makingtaskacrossaction
0
0 comments X
read the original abstract

Large vision-language models (VLMs) fine-tuned on specialized visual instruction-following data have exhibited impressive language reasoning capabilities across various scenarios. However, this fine-tuning paradigm may not be able to efficiently learn optimal decision-making agents in multi-step goal-directed tasks from interactive environments. To address this challenge, we propose an algorithmic framework that fine-tunes VLMs with reinforcement learning (RL). Specifically, our framework provides a task description and then prompts the VLM to generate chain-of-thought (CoT) reasoning, enabling the VLM to efficiently explore intermediate reasoning steps that lead to the final text-based action. Next, the open-ended text output is parsed into an executable action to interact with the environment to obtain goal-directed task rewards. Finally, our framework uses these task rewards to fine-tune the entire VLM with RL. Empirically, we demonstrate that our proposed framework enhances the decision-making capabilities of VLM agents across various tasks, enabling 7b models to outperform commercial models such as GPT4-V or Gemini. Furthermore, we find that CoT reasoning is a crucial component for performance improvement, as removing the CoT reasoning results in a significant decrease in the overall performance of our method.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Training Software Engineering Agents and Verifiers with SWE-Gym

    cs.SE 2024-12 conditional novelty 8.0

    SWE-Gym supplies 2438 executable real-world Python tasks to train SWE agents and verifiers, yielding up to 19% gains and new open-weight SOTA of 32% on SWE-Bench Verified.

  2. Reasoning Portability: Guiding Continual Learning for MLLMs in the RLVR Era

    cs.LG 2026-05 unverdicted novelty 7.0

    Formalizes Reasoning Portability (RP) and proposes RDB-CL to modulate per-sample KL regularization in RLVR for MLLM continual learning, achieving +12.0% Last accuracy over vanilla RLVR baseline by preserving reusable ...

  3. Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs

    cs.CV 2024-06 unverdicted novelty 7.0

    Cambrian-1 is a vision-centric multimodal LLM family that evaluates over 20 vision encoders, introduces CV-Bench and the Spatial Vision Aggregator, and releases open models, code, and data achieving strong performance...

  4. Rank-Then-Act: Reward-Free Control from Frame-Order Progress

    cs.LG 2026-07 unverdicted novelty 6.0

    RTA trains a VLM as a progress ordinal scorer via GRPO on shuffled expert frames and uses Spearman rank correlation with temporal indices as a bounded RL reward, matching or exceeding prior video reward methods on dis...

  5. Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation

    cs.CV 2026-06 unverdicted novelty 6.0

    Qwen-RobotWorld is a language-conditioned video world model using Double-Stream MMDiT, an 8.6M-frame embodied corpus, and progressive curriculum training that ranks first on EWMBench and DreamGen Bench.

  6. Grounded Reinforcement Learning for Visual Reasoning

    cs.CV 2025-05 unverdicted novelty 6.0

    ViGoRL introduces visually grounded RL that anchors reasoning steps to image coordinates and uses multi-turn zooming to outperform standard RL and supervised baselines on spatial and GUI reasoning benchmarks.

  7. PRISM: Perception Reasoning Interleaved for Sequential Decision Making

    cs.AI 2026-05 unverdicted novelty 5.0

    PRISM interleaves VLM perception and LLM reasoning via a dynamic goal-oriented question-answer pipeline to produce sharper scene descriptions, outperforming prior image-based models on ALFWorld and Room-to-Room.

  8. UAV-VL-R1: Generalizing Vision-Language Models via Supervised Fine-Tuning and Multi-Stage GRPO for UAV Visual Reasoning

    cs.CV 2025-08 unverdicted novelty 5.0

    UAV-VL-R1 combines SFT and multi-stage GRPO reinforcement learning on a new 50,019-sample HRVQA-VL dataset to deliver substantially higher zero-shot accuracy on UAV visual reasoning tasks than both its 2B baseline and...

  9. Large Language Model-Brained GUI Agents: A Survey

    cs.AI 2024-11 unverdicted novelty 4.0

    A survey consolidating frameworks, data practices, large action models, benchmarks, applications, and research gaps in LLM-brained GUI agents.