Revisit to Segment: Working Memory Distillation for Reasoning Segmentation

arXiv:2609.34863 · cs.CV · Submitted 2026-09-28 · Read on arXiv

cs.CV

Submitted: 2026-09-28

Updated: 2026-10-03

Code: https://github.com/yancilin/SWiM

Terminology

• Distillation — 78× in this paper · explained in 2 episodes
• Benchmark — 20× in this paper · explained in 2 episodes
• Reinforcement Learning — 19× in this paper · explained in 4 episodes
• GRPO — 14× in this paper · explained in 6 episodes
• Self-Distillation — 11× in this paper · explained in 7 episodes
• Trace — 10× in this paper · explained in 2 episodes
• On-Policy Distillation — 8× in this paper · explained in 3 episodes
• Reasoning Traces — 8× in this paper · explained in 3 episodes
• Core — 7× in this paper · explained in 2 episodes
• Conference on Computer Vision and Pattern Recognition — 4× in this paper
• Fine-Tuning — 4× in this paper · explained in 5 episodes
• Mask Generation — 4× in this paper · explained in 2 episodes
• Ablation — 3× in this paper · explained in 1 episode
• Ablation Studies — 2× in this paper · explained in 1 episode
• Generalization — 2× in this paper · explained in 13 episodes
• Grounding — 2× in this paper · explained in 3 episodes
• Regularization — 2× in this paper · explained in 3 episodes
• Reinforcement Learning (RL) — 2× in this paper · explained in 25 episodes
• Actor — 1× in this paper · explained in 3 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Alignment — 1× in this paper · explained in 1 episode
• Annotation Errors — 1× in this paper · explained in 1 episode
• Contextual Understanding — 1× in this paper · explained in 2 episodes
• GRPO Objective — 1× in this paper · explained in 2 episodes
• Group Relative Policy Optimization — 1× in this paper
• Group Relative Policy Optimization (GRPO) — 1× in this paper · explained in 5 episodes
• Multimodal Reasoning — 1× in this paper · explained in 5 episodes
• On-policy Self-Distillation (OPSD) — 1× in this paper · explained in 2 episodes
• Physical Constraints — 1× in this paper · explained in 7 episodes
• Structured Reasoning — 1× in this paper · explained in 3 episodes

Sources

• Qwen3-VL Technical Report
• Reason Twice: Segmentation via Candidate Discovery and Comparative Reasoning
• DR$^2$Seg: Decomposed Two-Stage Rollouts for Efficient Reasoning Segmentation in Multimodal Large Language Models
• LoRA: Low-Rank Adaptation of Large Language Models
• Rethinking On-Policy Distillation of Large Language Models: Phenomenology, Mechanism, and Recipe
• Seg-ReSearch: Segmentation with Interleaved Reasoning and External Search
• Seg-Zero: Reasoning-Chain Guided Segmentation via Cognitive Reinforcement
• From Failure to Feedback: Group Revision Unlocks Hard Cases in Object-Level Grounding
• Decoupled Weight Decay Regularization
• SegCompass: Exploring Interpretable Alignment with Sparse Autoencoders for Enhanced Reasoning Segmentation
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• A Survey of On-Policy Distillation for Large Language Models
• Gemini: A Family of Highly Capable Multimodal Models
• Kimi K3: Open Frontier Intelligence
• Skill-SD: Skill-Conditioned Self-Distillation for Multi-turn LLM Agents
• SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning
• Discriminative Perception via Anchored Description for Reasoning Segmentation
• Don't Guess, Just Ask: Resolving Ambiguity in Referring Segmentation via Multi-turn Clarification
• Multi-Rollout On-Policy Distillation via Peer Successes and Failures
• StAR: Segment Anything Reasoner

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models