When Words Outperform Vision: VLMs Can Self-Improve Via Text-Only Training For Human-Centered Decision Making
- URL: http://arxiv.org/abs/2503.16965v1
- Date: Fri, 21 Mar 2025 09:25:23 GMT
- Title: When Words Outperform Vision: VLMs Can Self-Improve Via Text-Only Training For Human-Centered Decision Making
- Authors: Zhe Hu, Jing Li, Yu Yin,
- Abstract summary: Embodied decision-making is fundamental for AI agents operating in real-world environments.<n>In this study, we evaluate open-sourced Visual Language Models (VLMs) on multimodal human-centered decision-making tasks.
- Score: 15.397582422113627
- License: http://creativecommons.org/licenses/by-nc-sa/4.0/
- Abstract: Embodied decision-making is fundamental for AI agents operating in real-world environments. While Visual Language Models (VLMs) have advanced this capability, they still struggle with complex decisions, particularly in human-centered situations that require deep reasoning about human needs and values. In this study, we systematically evaluate open-sourced VLMs on multimodal human-centered decision-making tasks. We find that LLMs receiving only textual descriptions unexpectedly outperform their VLM counterparts of similar scale that process actual images, suggesting that visual alignment may hinder VLM abilities. To address this challenge, we propose a novel text-only training approach with synthesized textual data. This method strengthens VLMs' language components and transfers the learned abilities to multimodal inference, eliminating the need for expensive image-text paired data. Furthermore, we show that VLMs can achieve substantial performance gains through self-improvement, using training data generated by their LLM counterparts rather than relying on larger teacher models like GPT-4. Our findings establish a more efficient and scalable approach to enhancing VLMs' human-centered decision-making capabilities, opening new avenues for optimizing VLMs through self-improvement mechanisms.
Related papers
- Weakly-supervised VLM-guided Partial Contrastive Learning for Visual Language Navigation [36.17444261325021]
Visual Language Navigation (VLN) is a fundamental task within the field of Embodied AI, focusing on the ability of agents to navigate complex environments based on natural language instructions.<n>Existing methods rely on pre-trained backbone models for visual perception, which struggle with the dynamic viewpoints in VLN scenarios.<n>We propose Weakly-supervised Partial Contrastive Learning (WPCL), a method that enhances an agent's ability to identify objects from dynamic viewpoints in VLN scenarios without requiring VLM fine-tuning.
arXiv Detail & Related papers (2025-06-18T11:43:50Z) - Perceptual Decoupling for Scalable Multi-modal Reasoning via Reward-Optimized Captioning [78.17782197231325]
We propose a reasoning-guided reinforcement learning strategy that aligns the extractor's captioning behavior with the reasoning objective.<n> Experiments on multi-modal math and science benchmarks show that the proposed RACRO method achieves state-of-the-art average performance.
arXiv Detail & Related papers (2025-06-05T02:28:07Z) - An Empirical Study of Federated Prompt Learning for Vision Language Model [50.73746120012352]
This paper systematically investigates behavioral differences between language prompt learning and vision prompt learning.<n>We conduct experiments to evaluate the impact of various fl and prompt configurations, such as client scale, aggregation strategies, and prompt length.<n>We explore strategies for enhancing prompt learning in complex scenarios where label skew and domain shift coexist.
arXiv Detail & Related papers (2025-05-29T03:09:15Z) - Decoupled Visual Interpretation and Linguistic Reasoning for Math Problem Solving [57.22004912994658]
Current large vision-language models (LVLMs) typically employ a connector module to link visual features with text embeddings of large language models (LLMs)<n>This paper proposes a paradigm shift: instead of training end-to-end vision-language reasoning models, we advocate for developing a decoupled reasoning framework.
arXiv Detail & Related papers (2025-05-23T08:18:00Z) - CausalVLBench: Benchmarking Visual Causal Reasoning in Large Vision-Language Models [10.530681458312412]
Large vision-language models (LVLMs) have shown impressive performance in tasks such as recognition and visual question answering.<n>We introduce a comprehensive causal reasoning benchmark for multi-modal in-context learning from LVLMs.<n>We evaluate the ability of state-of-the-art open-source LVLMs on our causal reasoning tasks across three causal representation learning datasets.
arXiv Detail & Related papers (2025-05-21T00:45:15Z) - VIPER: Visual Perception and Explainable Reasoning for Sequential Decision-Making [21.61801132083334]
VIPER is a novel framework for multimodal instruction-based planning.<n>It integrates VLM-based perception with LLM-based reasoning.<n>We show that VIPER significantly outperforms state-of-the-art visual instruction-based planners.
arXiv Detail & Related papers (2025-03-19T11:05:42Z) - Reflective Planning: Vision-Language Models for Multi-Stage Long-Horizon Robotic Manipulation [90.00687889213991]
Solving complex long-horizon robotic manipulation problems requires sophisticated high-level planning capabilities.
Vision-language models (VLMs) pretrained on Internet data could in principle offer a framework for tackling such problems.
In this paper, we introduce a novel test-time framework that enhancesVLMs' physical reasoning capabilities for multi-stage manipulation tasks.
arXiv Detail & Related papers (2025-02-23T20:42:15Z) - Preference VLM: Leveraging VLMs for Scalable Preference-Based Reinforcement Learning [17.59802090014789]
We introduce PrefVLM, a framework that integrates Vision-Language Models (VLMs) with selective human feedback.<n>Our method leverages VLMs to generate initial preference labels, which are then filtered to identify uncertain cases for targeted human annotation.<n> Experiments on Meta-World manipulation tasks demonstrate that PrefVLM achieves comparable or superior success rates to state-of-the-art methods.
arXiv Detail & Related papers (2025-02-03T18:50:15Z) - Improving Fine-grained Visual Understanding in VLMs through Text-Only Training [0.0]
We investigate the feasibility of enhancing fine-grained visual understanding in Visual-Language Models (VLMs) through text-only training.<n>We conduct comprehensive experiments on two distinct domains: fine-grained species classification and cultural visual understanding tasks.<n>Our findings demonstrate that text-only training can be comparable to conventional image-text training while significantly reducing computational costs.
arXiv Detail & Related papers (2024-12-17T14:18:50Z) - Enhancing Instruction-Following Capability of Visual-Language Models by Reducing Image Redundancy [37.471419716572086]
There is a significant gap in instruction-following capabilities between Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs)
We propose Visual-Modality Token Compression (VMTC) and Cross-Modality Attention Inhibition (CMAI) strategies to alleviate this gap.
arXiv Detail & Related papers (2024-11-23T05:03:32Z) - Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning [79.38140606606126]
We propose an algorithmic framework that fine-tunes vision-language models (VLMs) with reinforcement learning (RL)
Our framework provides a task description and then prompts the VLM to generate chain-of-thought (CoT) reasoning.
We demonstrate that our proposed framework enhances the decision-making capabilities of VLM agents across various tasks.
arXiv Detail & Related papers (2024-05-16T17:50:19Z) - Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing [56.75702900542643]
We introduce AlphaLLM for the self-improvements of Large Language Models.
It integrates Monte Carlo Tree Search (MCTS) with LLMs to establish a self-improving loop.
Our experimental results show that AlphaLLM significantly enhances the performance of LLMs without additional annotations.
arXiv Detail & Related papers (2024-04-18T15:21:34Z) - Causality-based Cross-Modal Representation Learning for
Vision-and-Language Navigation [15.058687283978077]
Vision-and-Language Navigation (VLN) has gained significant research interest in recent years due to its potential applications in real-world scenarios.
Existing VLN methods struggle with the issue of spurious associations, resulting in poor generalization with a significant performance gap between seen and unseen environments.
We propose a unified framework CausalVLN based on the causal learning paradigm to train a robust navigator capable of learning unbiased feature representations.
arXiv Detail & Related papers (2024-03-06T02:01:38Z) - MOKA: Open-World Robotic Manipulation through Mark-Based Visual Prompting [97.52388851329667]
We introduce Marking Open-world Keypoint Affordances (MOKA) to solve robotic manipulation tasks specified by free-form language instructions.
Central to our approach is a compact point-based representation of affordance, which bridges the VLM's predictions on observed images and the robot's actions in the physical world.
We evaluate and analyze MOKA's performance on various table-top manipulation tasks including tool use, deformable body manipulation, and object rearrangement.
arXiv Detail & Related papers (2024-03-05T18:08:45Z) - Exploring the Frontier of Vision-Language Models: A Survey of Current Methodologies and Future Directions [11.786387517781328]
Vision-Language Models (VLMs) are advanced models that can tackle more intricate tasks such as image captioning and visual question answering.
Our classification organizes VLMs into three distinct categories: models dedicated to vision-language understanding, models that process multimodal inputs to generate unimodal (textual) outputs and models that both accept and produce multimodal inputs and outputs.
We meticulously dissect each model, offering an extensive analysis of its foundational architecture, training data sources, as well as its strengths and limitations wherever possible.
arXiv Detail & Related papers (2024-02-20T18:57:34Z) - LMRL Gym: Benchmarks for Multi-Turn Reinforcement Learning with Language
Models [56.25156596019168]
This paper introduces the LMRL-Gym benchmark for evaluating multi-turn RL for large language models (LLMs)
Our benchmark consists of 8 different language tasks, which require multiple rounds of language interaction and cover a range of tasks in open-ended dialogue and text games.
arXiv Detail & Related papers (2023-11-30T03:59:31Z) - Sight Beyond Text: Multi-Modal Training Enhances LLMs in Truthfulness
and Ethics [32.123919380959485]
Multi-modal large language models (MLLMs) are trained based on large language models (LLM)
While they excel in multi-modal tasks, the pure NLP abilities of MLLMs are often underestimated and left untested.
We show that visual instruction tuning, a prevailing strategy for transitioning LLMs into MLLMs, unexpectedly and interestingly helps models attain both improved truthfulness and ethical alignment.
arXiv Detail & Related papers (2023-09-13T17:57:21Z) - Measuring and Improving Chain-of-Thought Reasoning in Vision-Language Models [61.28463542324576]
Vision-language models (VLMs) have recently demonstrated strong efficacy as visual assistants that can generate human-like outputs.
We evaluate existing state-of-the-art VLMs and find that even the best-performing model is unable to demonstrate strong visual reasoning capabilities and consistency.
We propose a two-stage training framework aimed at improving both the reasoning performance and consistency of VLMs.
arXiv Detail & Related papers (2023-09-08T17:49:44Z) - TouchStone: Evaluating Vision-Language Models by Language Models [91.69776377214814]
We propose an evaluation method that uses strong large language models as judges to comprehensively evaluate the various abilities of LVLMs.
We construct a comprehensive visual dialogue dataset TouchStone, consisting of open-world images and questions, covering five major categories of abilities and 27 subtasks.
We demonstrate that powerful LVLMs, such as GPT-4, can effectively score dialogue quality by leveraging their textual capabilities alone.
arXiv Detail & Related papers (2023-08-31T17:52:04Z) - PEVL: Position-enhanced Pre-training and Prompt Tuning for
Vision-language Models [127.17675443137064]
We introduce PEVL, which enhances the pre-training and prompt tuning of vision-language models with explicit object position modeling.
PEVL reformulates discretized object positions and language in a unified language modeling framework.
We show that PEVL enables state-of-the-art performance on position-sensitive tasks such as referring expression comprehension and phrase grounding.
arXiv Detail & Related papers (2022-05-23T10:17:53Z)
This list is automatically generated from the titles and abstracts of the papers in this site.
This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.