AI Paper Atlas

Research map / Vision-language and multimodal / Multimodal LLMs

SAM-Veteran: An MLLM-Based Human-like SAM Agent for Reasoning Segmentation

Tianyuan Du, Haopeng Li, Zhen Fan, Jiarui Zhang, Panwang Pan, Yang Zhang

ICLR 2026 · Vision-language and multimodal › Multimodal LLMs · approach: Chain-of-thought multimodal reasoning

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes SAM-Veteran, an MLLM-based agent that uses SAM as an interactive tool for iterative mask refinement in reasoning segmentation.

Problem

Existing MLLM+SAM reasoning segmentation paradigms do not exploit SAM's ability to support iterative, interactive mask refinement.

Approach

An MLLM agent that behaves more like a human annotator, interacting with SAM iteratively to refine masks rather than producing a single prompt-to-mask pass.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add mask comprehension task to RL training (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 70.3 -> 72.1Improved
add auxiliary corrupted-mask task (ablation)SAM-Veteran · 5-benchmark averageavg IoU; termination: 72.1 -> 72.2; Never -> Adaptive terminationImproved
add textual grounding RL training to Qwen2.5-VL (ablation)Qwen+SAM2 · 5-benchmark averageavg IoU: 64.4 -> 70.3Improved
add SAM reward (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 71.3 -> 72.2Improved
add decision reward (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 70.6 -> 72.2Improved
add IoU improvement reward (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 70.4 -> 72.2Improved
use graded IoU improvement reward instead of hard version (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 71.9 -> 72.2Improved
add dynamic sampling (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 70.8 -> 72.2Improved
add chain of thought in all tasks (ablation)SAM-Veteran · ReasonSeg valgIoU: 64.4 -> 68.2Improved
add chain of thought in mask tasks (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 71.6 -> 72.2Improved
iterative refinement from the trained model vs QwenSAM-Veteran · ReasonSeg valgIoU change over steps: 0.00 -> 2.47 (Qwen: 0.00 -> -2.48)Improved
use hard IoU box reward threshold 0.5 instead of soft IoU reward (ablation)SAM-Veteran · 5-benchmark averageavg IoU: 70.7 -> 72.3Improved

Later papers that compared with this method

1 later accepted paper reports SAM-Veteran as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Experimental setup

Builds on: Segment Anything Model

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy