REVIEW 25 cited by
LISA++: An Improved Baseline for Reasoning Segmentation with Large Language Model
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
Signed reviews
read the original abstract
While LISA effectively bridges the gap between segmentation and large language models to enable reasoning segmentation, it poses certain limitations: unable to distinguish different instances of the target region, and constrained by the pre-defined textual response formats. In this work, we introduce LISA++, an update to the existing LISA model, focusing on improving core functionalities while keeping the base architecture intact. The main enhancements in LISA++ include: \textbf{1) Enhanced Segmentation}: The instance segmentation ability has been added, providing a more detailed scene analysis along with the existing multi-region semantic segmentation. \textbf{2) More Natural Conversation}: Improved capability for multi-turn dialogue, with the ability to incorporate segmentation results directly into text responses, i.e., Segmentation in Dialogue (SiD). These improvements are achieved by curating the existing samples of generic segmentation datasets, aimed specifically at enhancing the segmentation and conversational skills without structural change and additional data sources. Comparative analysis with the original LISA model shows significant advancements in these areas, positioning LISA++ as a notable upgrade in visual understanding and interaction. LISA++'s adaptability and improved features highlight the versatility of the mask-as-embedding paradigm proposed by LISA, and the potential as a foundational model for diverse applications.
Forward citations
Cited by 25 Pith papers
-
Vision as Unified Multimodal Generation
A single unified multimodal model matches leading task-specialized vision systems across detection, segmentation, dense geometry, and multi-view 3D by casting all outputs as native text or image generation.
-
PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation
The work introduces the UAV Reasoning Segmentation task, the DRSeg benchmark dataset, and PixDLM as a baseline dual-path multimodal language model for reasoning-based segmentation in aerial imagery.
-
Vision-Language Grounding as Bidirectional Concept Correspondence
Grounding can be treated as recovering all word-span to image-mask correspondences in one pass, and a bridge-token model, ConCor-1, does this better than existing grounding pipelines on the tested benchmarks.
-
Scaling Representation Diversity: Modulated Attention and Reconstructive Regularization for Visual Grounding
A unified visual grounding framework combining a broadcast cross-attention head, a JEPA auxiliary loss, and an MLLM-generated caption dataset preserves representation diversity and generalizes across RefCOCO/+/g.
-
MedPixel: A Unified Pixel-Language Model for Medical Reasoning and Segmentation
A unified medical pixel-language model trained on 440K synthesized mask-language samples achieves strong performance on reasoning and explanatory segmentation, with zero-shot transfer to external grounding benchmarks.
-
RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection
Depth refinement for monocular 3D detection is reformulated as a visual alignment judgment made by a VLM emitting discrete direction and magnitude tokens, yielding consistent post-hoc gains without retraining the base...
-
ReferTrack: Referring Then Tracking for Embodied Visual Tracking
A refer-then-track policy picks the target from indexed detections before planning waypoints, achieving state-of-the-art single-view results on EVT-Bench and approaching multi-camera performance.
-
Reasoning-Guided Part-Level Visual Grounding via Reinforcement Learning
A 4B MLLM trained with object-first hierarchical prompting and stage-wise part-aware GRPO rewards outperforms 7B grounding LLMs and SAM3 on part-grounding benchmarks.
-
SCOPE: Speech-guided COllaborative PErception Framework for Surgical Scene Segmentation
A speech-guided framework uses an LLM and open-set vision models to segment and track surgical instruments and anatomy hands-free in live video.
-
Object-centric Video Question Answering with Visual Grounding and Referring
RGA3 unifies visual referring (arbitrary prompts at any timestamp) and grounding (segmentation masks) for object-centric video QA, introducing the STOM prompt-propagation module and the VideoInfer dataset.
-
VisionThink: Smart and Efficient Vision Language Model via Reinforcement Learning
A vision-language model learns via reinforcement learning when to upscale a low-resolution image, cutting visual tokens roughly in half while preserving accuracy on most benchmarks.
-
HRSeg: High-Resolution Visual Perception and Enhancement for Reasoning Segmentation
HRSeg combines high-resolution image crops, region attention, and cross-attention mask enhancement, improving reasoning segmentation over LLM-Seg by up to 13.2 gIoU points on LLM-Seg40K.
-
LIRA: Inferring Segmentation in Large Multi-modal Models with Local Interleaved Region Assistance
LIRA improves referring segmentation and reduces hallucination in multimodal LLMs by fusing semantic and pixel features and interleaving local image regions with text descriptions.
-
Enhancing Spatial Reasoning in Multimodal Large Language Models through Reasoning-based Segmentation
A two-stage reasoning-segmentation method plus a new LLM-generated 3D dataset improves spatial reasoning in 3D multimodal large language models on several benchmarks.
-
R2SM: Referring and Reasoning for Selective Masks
R2SM provides the first benchmark pairing modal and amodal text prompts with matching masks, letting models learn when to segment only visible parts versus complete occluded shapes.
-
PixelThink: Towards Efficient Chain-of-Pixel Reasoning
A soft token-budget penalty driven by task difficulty and model confidence roughly halves the reasoning tokens used by segmentation MLLMs while slightly improving mask accuracy on ReasonSeg-derived benchmarks.
-
TrackVLA: Embodied Visual Tracking in the Wild
A single vision-language-action model jointly trained on recognition and tracking data follows described targets at the best reported levels on a public benchmark and transfers zero-shot from simulation to a real quad...
-
DyFo: A Training-Free Dynamic Focus Visual Search for Enhancing LMMs in Fine-Grained Visual Understanding
DyFo improves fine-grained visual understanding in large multimodal models by using MCTS-guided interaction with a visual expert to focus on relevant image regions, without additional training.
-
HandsOnVLM: Vision-Language Models for Hand-Object Interaction Prediction
HandsOnVLM predicts future 2D hand trajectories from egocentric video and language instructions by treating trajectories as auto-regressive <HAND> tokens in a vision-language model.
-
InsightEdit: Towards Better Instruction Following for Image Editing
InsightEdit uses multimodal language model features in a two-stream adapter to improve complex instruction following and background consistency in image editing.
-
StAR: Segment Anything Reasoner
StAR, a low-rank RL fine-tuning recipe with mask-aware rewards, selective rollout expansion, and label-before-box answers, lifts reasoning-segmentation gIoU by roughly 7 points and adds a new four-type benchmark, ReasonSeg-X.
-
MediRound: Multi-Round Entity-Level Reasoning Segmentation in Medical Images
MediRound introduces a multi-round, entity-level medical segmentation task, a 177K-dialogue dataset built from SA-Med2D-20M with GPT-5, and a LLaVA-Med/MedSAM baseline whose inference-time judgment-and-correction modu...
-
Mask-aware Text-to-Image Retrieval: Referring Expression Segmentation Meets Cross-modal Retrieval
A zero-shot pipeline combining SAM2 masks, Alpha-CLIP region embeddings, and Qwen2.5-VL reranking and grounding improves mask-aware text-to-image retrieval on COCO and D3.
-
GLOVER++: Unleashing the Potential of Affordance Learning from Human Behaviors for Robotic Manipulation
GLOVER++ predicts where to interact with objects from language instructions after training on the new 500K-image HOVA-500K affordance dataset, and reports gains in simulated and real manipulation.
-
EditScout: Locating Forged Regions from Diffusion-based Edited Images with Multimodal LLM
A multimodal LLM with a SAM-based mask decoder localizes diffusion-edited regions better than traditional forensic methods on MagicBrush, CocoGLIDE, and a new BrushNet dataset.
Discussion (0). Continue with ORCID to comment.