Introduces APRS task and PanoSeeker agent using VLM plus EgoSphere memory for active 360° search and segmentation, outperforming baselines on a new benchmark.
hub Canonical reference
Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement
Canonical reference. 78% of citing Pith papers cite this work as background.
53
Pith papers citing it
Background
78% of classified citations
abstract
Traditional methods for reasoning segmentation rely on supervised fine-tuning with categorical labels and simple descriptions, limiting its out-of-domain generalization and lacking explicit reasoning processes. To address these limitations, we propose Seg-Zero, a novel framework that demonstrates remarkable generalizability and derives explicit chain-of-thought reasoning through cognitive reinforcement. Seg-Zero introduces a decoupled architecture consisting of a reasoning model and a segmentation model. The reasoning model interprets user intentions, generates explicit reasoning chains, and produces positional prompts, which are subsequently used by the segmentation model to generate precious pixel-level masks. We design a sophisticated reward mechanism that integrates both format and accuracy rewards to effectively guide optimization directions. Trained exclusively via reinforcement learning with GRPO and without explicit reasoning data, Seg-Zero achieves robust zero-shot generalization and exhibits emergent test-time reasoning capabilities. Experiments show that Seg-Zero-7B achieves a zero-shot performance of 57.5 on the ReasonSeg benchmark, surpassing the prior LISA-7B by 18\%. This significant improvement highlights Seg-Zero's ability to generalize across domains while presenting an explicit reasoning process.
hub tools
citation-role summary
background 7
baseline 2
citation-polarity summary
representative citing papers
P2R decouples perception from reasoning in VLMs via a two-stage process and PRA-GRPO alternating RL training, reporting gains such as 93.2% on V-Star for the 4B model over its Qwen3-VL backbone.
Introduces the FeVOS task, a 968-clip dataset with foresight expressions, and an MLLM model FeVOS-R1 trained via SFT then RL that reports SOTA on the new task plus generalization to prior RVOS benchmarks.
Rea2Seg turns image segmentation into candidate mask discovery from MLLM attention followed by MLLM-based comparative scoring and selection, plus a new multi-dimensional reasoning benchmark ReasonSeg-SGDR.
VAMPS benchmark shows multimodal LLMs solve graph-assisted math problems better by direct analysis than by tool-enabled visual solving.
MAOAM unifies object and material selection via a VLM with segmentation head, supporting text and click interactions through multi-task training on VLM-generated material data.
VisHarness learns a reinforcement-learned policy to harness specialized visual experts via multi-turn interactions and dynamic visual memory archiving, outperforming general models on four visual reasoning benchmarks.
ConceptSeg-R1 uses Meta-GRPO meta-RL to learn transferable rules from visual demonstrations and apply them via concept translation for generalized concept segmentation across CI, CD, and CR levels.
SetCon achieves state-of-the-art open-ended referring segmentation by using LVLM-generated set-level concepts for joint mask decoding, with gains increasing for multi-target cases on image and video benchmarks.
VASA is a vision-guided agent for open ad-hoc segmentation that creates and validates masks through planning, tool use, and error recovery, outperforming baselines on the new PARS benchmark and RefCOCOm.
Seg-Agent performs language-guided segmentation without training by using Set-of-Mark visual prompts to enable explicit multimodal chain-of-reasoning in three stages: generation, selection, and refinement.
WebEye benchmark and Pixel-Searcher agent enable visual perception tasks by using web search to resolve object identities before precise localization or answering.
An image generator is instruction-tuned to perform diverse vision tasks by representing task outputs as RGB images, achieving SOTA on segmentation and depth estimation.
Topo-R1 fine-tunes a vision-language model using a topology-aware reward and GRPO to detect anomalies such as broken or spurious connections in tubular segmentation masks, outperforming standard VLMs.
LASAGNA produces layered images with integrated visual effects in a single pass, enabling drift-free edits via alpha compositing while releasing a 48K dataset and a 242-sample benchmark.
IBISAgent enables MLLMs to perform iterative pixel-level visual reasoning for biomedical object referring and segmentation via text-based clicks and agentic RL, outperforming prior SOTA methods without model modifications.
SAM 3 introduces promptable concept segmentation that doubles accuracy of prior systems on images and videos while improving standard SAM segmentation performance.
Proposes CSR task and HalluSegBench using visual counterfactuals to diagnose segmentation hallucinations in VLMs, plus RobustSeg via counterfactual fine-tuning that reduces hallucinations by 30% on FP-RefCOCO.
Chain-of-Zoom factorizes extreme super-resolution into an autoregressive sequence of intermediate scales using a reused backbone model plus GRPO-tuned multi-scale VLM prompts.
DeepEyes uses reinforcement learning to teach vision-language models active perception and image-based thinking, yielding gains on perception, reasoning, grounding, and hallucination benchmarks.
InstanceControl uses VLMs to auto-generate instance masks from text and visual conditions, with adaptive refinement, to enable controllable multi-object image generation without manual labeling.
DR-MV3D decomposes MV3D-VQA into global map construction, question-conditioned view planning, and egocentric grounding, supervised by global consistency and local trajectory rewards optimized via GRPO.
Attention maps in LVLMs enable an IoU regressor (Pearson r > 0.67) and a training-free entropy-based selector that improves small-object localization by up to 19% on COCO and Objects365.
Introduces MTRS task, MTRefSeg-21K benchmark of 21K image-text-mask triplets, and MTRefSeg-R1 LVLM baseline that outperforms standard models via two-stage change-aware training.
citing papers explorer
-
Seek to Segment: Active Perception for Panoramic Referring Segmentation
Introduces APRS task and PanoSeeker agent using VLM plus EgoSphere memory for active 360° search and segmentation, outperforming baselines on a new benchmark.
-
Perceive-to-Reason: Decoupling Perception and Reasoning for Fine-Grained Visual Reasoning
P2R decouples perception from reasoning in VLMs via a two-stage process and PRA-GRPO alternating RL training, reporting gains such as 93.2% on V-Star for the 4B model over its Qwen3-VL backbone.
-
FeVOS: Foresight Expression Video Object Segmentation
Introduces the FeVOS task, a 968-clip dataset with foresight expressions, and an MLLM model FeVOS-R1 trained via SFT then RL that reports SOTA on the new task plus generalization to prior RVOS benchmarks.
-
Reason Twice: Segmentation via Candidate Discovery and Comparative Reasoning
Rea2Seg turns image segmentation into candidate mask discovery from MLLM attention followed by MLLM-based comparative scoring and selection, plus a new multi-dimensional reasoning benchmark ReasonSeg-SGDR.
-
VAMPS: Visual-Assisted Mathematical Problem Solving Benchmark
VAMPS benchmark shows multimodal LLMs solve graph-assisted math problems better by direct analysis than by tool-enabled visual solving.
-
MAOAM: Unified Object and Material Selection with Vision-Language Models
MAOAM unifies object and material selection via a VLM with segmentation head, supporting text and click interactions through multi-task training on VLM-generated material data.
-
Train the Agent, Not the Expert: Learning to Harness Heterogeneous Experts for Multi-Turn Visual Reasoning
VisHarness learns a reinforcement-learned policy to harness specialized visual experts via multi-turn interactions and dynamic visual memory archiving, outperforming general models on four visual reasoning benchmarks.
-
ConceptSeg-R1: Segment Any Concept via Meta-Reinforcement Learning
ConceptSeg-R1 uses Meta-GRPO meta-RL to learn transferable rules from visual demonstrations and apply them via concept translation for generalized concept segmentation across CI, CD, and CR levels.
-
SetCon: Towards Open-Ended Referring Segmentation via Set-Level Concept Prediction
SetCon achieves state-of-the-art open-ended referring segmentation by using LVLM-generated set-level concepts for joint mask decoding, with gains increasing for multi-target cases on image and video benchmarks.
-
Vision Harnessing Agent for Open Ad-hoc Segmentation
VASA is a vision-guided agent for open ad-hoc segmentation that creates and validates masks through planning, tool use, and error recovery, outperforming baselines on the new PARS benchmark and RefCOCOm.
-
Seg-Agent: Test-Time Multimodal Reasoning for Training-Free Language-Guided Segmentation
Seg-Agent performs language-guided segmentation without training by using Set-of-Mark visual prompts to enable explicit multimodal chain-of-reasoning in three stages: generation, selection, and refinement.
-
From Web to Pixels: Bringing Agentic Search into Visual Perception
WebEye benchmark and Pixel-Searcher agent enable visual perception tasks by using web search to resolve object identities before precise localization or answering.
-
Image Generators are Generalist Vision Learners
An image generator is instruction-tuned to perform diverse vision tasks by representing task outputs as RGB images, achieving SOTA on segmentation and depth estimation.
-
Topo-R1: Detecting Topological Anomalies via Vision-Language Models
Topo-R1 fine-tunes a vision-language model using a topology-aware reward and GRPO to detect anomalies such as broken or spurious connections in tubular segmentation masks, outperforming standard VLMs.
-
A Unified and Controllable Framework for Layered Image Generation with Visual Effects
LASAGNA produces layered images with integrated visual effects in a single pass, enabling drift-free edits via alpha compositing while releasing a 48K dataset and a 242-sample benchmark.
-
IBISAgent: Reinforcing Pixel-Level Visual Reasoning in MLLMs for Universal Biomedical Object Referring and Segmentation
IBISAgent enables MLLMs to perform iterative pixel-level visual reasoning for biomedical object referring and segmentation via text-based clicks and agentic RL, outperforming prior SOTA methods without model modifications.
-
SAM 3: Segment Anything with Concepts
SAM 3 introduces promptable concept segmentation that doubles accuracy of prior systems on images and videos while improving standard SAM segmentation performance.
-
Counterfactual Segmentation Reasoning: Diagnosing and Mitigating Pixel-Grounding Hallucination
Proposes CSR task and HalluSegBench using visual counterfactuals to diagnose segmentation hallucinations in VLMs, plus RobustSeg via counterfactual fine-tuning that reduces hallucinations by 30% on FP-RefCOCO.
-
Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment
Chain-of-Zoom factorizes extreme super-resolution into an autoregressive sequence of intermediate scales using a reused backbone model plus GRPO-tuned multi-scale VLM prompts.
-
DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning
DeepEyes uses reinforcement learning to teach vision-language models active perception and image-based thinking, yielding gains on perception, reasoning, grounding, and hallucination benchmarks.
-
InstanceControl: Controllable Complex Image Generation without Instance Labeling
InstanceControl uses VLMs to auto-generate instance masks from text and visual conditions, with adaptive refinement, to enable controllable multi-object image generation without manual labeling.
-
Dense Reward for Multi-View 3D Reasoning with Global Maps and Local Views
DR-MV3D decomposes MV3D-VQA into global map construction, question-conditioned view planning, and egocentric grounding, supervised by global consistency and local trajectory rewards optimized via GRPO.
-
Self-Improving Small Object Grounding in LVLMs
Attention maps in LVLMs enable an IoU regressor (Pearson r > 0.67) and a training-free entropy-based selector that improves small-object localization by up to 19% on COCO and Objects365.
-
An Open-Source Benchmark and Baseline for Multi-temporal Referring Segmentation
Introduces MTRS task, MTRefSeg-21K benchmark of 21K image-text-mask triplets, and MTRefSeg-R1 LVLM baseline that outperforms standard models via two-stage change-aware training.
-
MedVol-R1: Reward-Driven Evidence Grounding for Volumetric Reasoning Segmentation
MedVol-R1 is an RL framework that decouples 2D evidence grounding from 3D mask generation for volumetric reasoning segmentation and reports SOTA results on M3D-Seg benchmarks.
-
VersusQ: Pairwise Margin Reasoning for Generalizable Video Quality Assessment
VersusQ introduces a pairwise margin reasoning framework using large multimodal models to predict signed continuous quality margins between video pairs, claiming improved cross-domain generalization over pointwise scoring methods.
-
Are Tools Always Beneficial? Learning to Invoke Tools Adaptively for Dual-Mode Multimodal LLM Reasoning
AutoTool uses dual-mode RL to let MLLMs adaptively choose tool use or text-only reasoning, reporting 21.8% accuracy gain on V* and 44.9% efficiency gain on POPE versus baselines.
-
Don't Guess, Just Ask: Resolving Ambiguity in Referring Segmentation via Multi-turn Clarification
IC-Seg is a multi-turn clarification framework with hierarchical GRPO optimization that resolves ambiguous queries in referring video object segmentation and introduces the Ambi-RVOS benchmark.
-
From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding
A group-revision paradigm for GRPO-based RL fine-tuning of VLMs converts failure responses into improvement signals that refine rewards and advantages, yielding gains on referring segmentation, REC, and counting benchmarks.
-
Affordance Agent Harness: Verification-Gated Skill Orchestration
Affordance Agent Harness is a verification-gated orchestration system that unifies skills via an evidence store, episodic memory priors, an adaptive router, and a self-consistency verifier to improve accuracy-cost tradeoffs in open-world affordance grounding.
-
Video-ToC: Video Tree-of-Cue Reasoning
Video-ToC adds tree-guided cue localization, demand-based RL rewards, and automated datasets to video LLMs, reporting better results than prior methods on six understanding benchmarks plus a hallucination test.
-
PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation
The paper defines UAV reasoning segmentation, releases DRSeg—10,000 aerial images with reasoning QA and masks—and shows its PixDLM baseline beats prior reasoning-segmentation models under fine-tuning.
-
Saliency-R1: Enforcing Interpretable and Faithful Vision-language Reasoning via Saliency-map Alignment Reward
Saliency-R1 uses a novel saliency map technique and GRPO with human bounding-box overlap as reward to improve VLM reasoning faithfulness and interpretability.
-
AdaTooler-V: Adaptive Tool-Use for Images and Videos
AdaTooler-V trains MLLMs to adaptively use vision tools via AT-GRPO reinforcement learning and new datasets, reaching 89.8% on V* and outperforming GPT-4o.
-
Skyra: AI-Generated Video Detection via Grounded Artifact Reasoning
Skyra is an MLLM that detects AI-generated videos by identifying and reasoning over grounded visual artifacts, supported by a new annotated dataset and benchmark.
-
LongVT: Incentivizing "Thinking with Long Videos" via Native Tool Calling
LongVT adds native video-cropping tool calling to LMMs for interleaved multimodal chain-of-tool-thought reasoning on long videos and releases VideoSIAH data for training and evaluation.
-
ViSurf: Visual Supervised-and-Reinforcement Fine-Tuning for Large Vision-and-Language Models
ViSurf unifies SFT and RLVR for LVLMs in one training stage by injecting ground-truth labels into rollouts and applying novel reward controls, outperforming standalone and two-stage baselines on diverse benchmarks.
-
From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Models
The survey formalizes MLLM perception as a unified vision-language capability and traces its evolution via a new five-stage taxonomy while outlining future challenges.
-
EARL: Towards a Unified Analysis-Guided Reinforcement Learning Framework for Egocentric Interaction Reasoning and Pixel Grounding
EARL uses analysis-guided RL with a two-stage parsing and AFS module to achieve 65.48% cIoU in pixel grounding on Ego-IRGBench, outperforming prior RL methods.
-
ARGUS: Policy-Adaptive Ad Governance via Evolving Reinforcement with Adversarial Umpiring
ARGUS uses a Prosecutor-Defender-Umpire multi-agent setup plus RAG and chain-of-thought rewards to adapt ad policy enforcement to new regulations using minimal fresh labels.
-
SignReasoner: Compositional Reasoning for Complex Traffic Sign Understanding via Functional Structure Units
SignReasoner decomposes traffic signs into functional structure units and uses a two-stage VLM post-training pipeline to achieve state-of-the-art compositional reasoning on a new benchmark.
-
Learning to Focus and Precise Cropping: A Reinforcement Learning Framework with Information Gaps and Grounding Loss for MLLMs
A two-stage RL method with information gaps and grounding loss trains MLLMs to focus on and precisely crop relevant image regions, yielding SOTA results on high-resolution VQA benchmarks.
-
Grounding Everything in Tokens for Multimodal Large Language Models
GETok partitions images with grid tokens and refines locations via offset tokens to enable better native 2D spatial reasoning in MLLMs.
-
OneThinker: All-in-one Reasoning Model for Image and Video
OneThinker unifies image and video reasoning in one model across 10 tasks via a 600k corpus, CoT-annotated SFT, and EMA-GRPO reinforcement learning, reporting strong results on 31 benchmarks plus some cross-task transfer.
-
Geo-R1: Improving Few-Shot Geospatial Referring Expression Understanding with Reinforcement Fine-Tuning
Geo-R1 uses reasoning-centric reinforcement fine-tuning to improve few-shot performance and generalization in geospatial referring expression understanding over supervised baselines.
-
RealSR-R1: Reinforcement Learning for Real-World Image Super-Resolution with Vision-Language Chain-of-Thought
RealSR-R1 introduces VLCoT-GRPO with four rewards to add understanding and reasoning to real-world image super-resolution models.
-
VideoChat-R1: Enhancing Spatio-Temporal Perception via Reinforcement Fine-Tuning
Reinforcement fine-tuning with temporal rewards produces VideoChat-R1, a video MLLM showing large gains on spatio-temporal perception benchmarks such as +31.8 temporal grounding and +31.2 object tracking.
-
B-GRTO: Bootstrapped Group Relative Tool Optimization for Referring Segmentation
B-GRTO pre-trains a segmentation tool via bootstrapped group relative optimization on GRPO rollouts, yielding substantial gains over plain GRPO on referring segmentation benchmarks.
-
RCoT-Seg: Reinforced Chain-of-Thought for Video Reasoning and Segmentation
RCoT-Seg uses GRPO-reinforced keyframe selection from a CoT-start corpus followed by SAM2 mask propagation to improve video object segmentation under implicit temporal instructions over prior MLLM sampling methods.
-
Improving the Reasoning of Multi-Image Grounding in MLLMs via Reinforcement Learning
A pipeline of chain-of-thought data synthesis, LoRA-based supervised fine-tuning, rejection sampling, and rule-based reinforcement learning raises multi-image grounding accuracy by 9.04% on MIG-Bench and 4.41% on average across seven other benchmarks.