MindEdit-Bench introduces six spatial reasoning tasks from 120 private indoor photo triplets, with two new counterfactual editing tasks where VLMs score 8-31% against 81-97% human accuracy.
Robobrain 2.5: Depth in sight, time in mind
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
years
2026 14representative citing papers
OVO-S-Bench provides 1680 human-annotated questions on 348 videos to measure streaming spatial intelligence in MLLMs across instantaneous perception, spatiotemporal tracking, spatial simulation, and allocentric mapping.
VLMs exhibit consistent vertical-distance entanglement in embeddings from perspective bias in natural images, producing accuracy gaps that a new synthetic benchmark SpatialTunnel exposes as model-intrinsic.
SceneFunRI benchmark shows current VLMs struggle severely with inferring locations of invisible functional objects, with the strongest model (Gemini 3 Flash) reaching only 15.20 CAcc@75.
A zero-shot MLLM pipeline for last-mile navigation that selects a view, grounds a task affordance, lifts it to 3D, and predicts a collision-aware base pose, beating MoTo by 3.13 points on OVMM.
3D HAMSTER adds depth encoding and reconstruction to VLMs to produce 3D waypoint sequences that feed directly into pointcloud policies, claiming better generalization than 2D baselines under shifts.
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
Introduces a new diagnostic benchmark and million-scale reasoning corpus showing that training on explicit causal traces improves next-state prediction in embodied planning, with reported gains from data scaling.
A co-evolutionary VLM-VGM loop on 500 unlabeled images raises planner success by 30 points and simulator success by 48 percent while beating fully supervised baselines.
LoHo-Manip enables robust long-horizon robot manipulation by using a receding-horizon VLM manager to output progress-aware subtask sequences and 2D visual traces that condition a VLA executor for automatic replanning.
A dual-system framework with a structured subtask interface, event-balanced training, and inference harness enables VLM-guided long-horizon robotic manipulation, achieving 95.5% on LIBERO-Long and 65% on real-world chemistry tasks.
ERQA-Plus is a new benchmark dataset with a structured taxonomy for evaluating fine-grained embodied reasoning in AI agents across perceptual, action-centric, social, navigation, and commonsense categories.
An 8B embodied foundation model trained on 15B tokens with multi-task RL and a Planner-Grounder-Corrector loop claims SOTA on 16/24 embodied VLM benchmarks and strong VLA/real-robot transfer.
RoboSpatialBrain applies forced <think> prefix activation and reference-frame redirection to RoboBrain2.5-8B-NV, achieving first place with 80.9% success on RoboSpatial-Home.
citing papers explorer
-
MindEdit-Bench: Benchmarking Object-Level Counterfactual Spatial Reasoning in VLMs from In-the-Wild Photos
MindEdit-Bench introduces six spatial reasoning tasks from 120 private indoor photo triplets, with two new counterfactual editing tasks where VLMs score 8-31% against 81-97% human accuracy.
-
OVO-S-Bench: A Hierarchical Benchmark for Streaming Spatial Intelligence in Multimodal LLMs
OVO-S-Bench provides 1680 human-annotated questions on 348 videos to measure streaming spatial intelligence in MLLMs across instantaneous perception, spatiotemporal tracking, spatial simulation, and allocentric mapping.
-
Why Far Looks Up: Probing Spatial Representation in Vision-Language Models
VLMs exhibit consistent vertical-distance entanglement in embeddings from perspective bias in natural images, producing accuracy gaps that a new synthetic benchmark SpatialTunnel exposes as model-intrinsic.
-
SceneFunRI: Reasoning the Invisible for Task-Driven Functional Object Localization
SceneFunRI benchmark shows current VLMs struggle severely with inferring locations of invisible functional objects, with the strongest model (Gemini 3 Flash) reaching only 15.20 CAcc@75.
-
UniLM-Nav: A Unified Framework for Zero-Shot Last-Mile Navigation
A zero-shot MLLM pipeline for last-mile navigation that selects a view, grounds a task affordance, lifts it to 3D, and predicts a collision-aware base pose, beating MoTo by 3.13 points on OVMM.
-
3D HAMSTER: Bridging Planning and Control in Hierarchical Vision Language Action Models through 3D Trajectory Guidance
3D HAMSTER adds depth encoding and reconstruction to VLMs to produce 3D waypoint sequences that feed directly into pointcloud policies, claiming better generalization than 2D baselines under shifts.
-
Vesta: A Generalist Embodied Reasoning Model
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
-
Token Predictors Are Not Planners: Building Physically Grounded Causal Reasoners
Introduces a new diagnostic benchmark and million-scale reasoning corpus showing that training on explicit causal traces improves next-state prediction in embodied planning, with reported gains from data scaling.
-
RoboEvolve: Co-Evolving Planner-Simulator for Robotic Manipulation with Limited Data
A co-evolutionary VLM-VGM loop on 500 unlabeled images raises planner success by 30 points and simulator success by 48 percent while beating fully supervised baselines.
-
Long-Horizon Manipulation via Trace-Conditioned VLA Planning
LoHo-Manip enables robust long-horizon robot manipulation by using a receding-horizon VLM manager to output progress-aware subtask sequences and 2D visual traces that condition a VLA executor for automatic replanning.
-
Cortex: A Bidirectionally Aligned Embodied Agent Framework for Long-horizon Manipulation
A dual-system framework with a structured subtask interface, event-balanced training, and inference harness enables VLM-guided long-horizon robotic manipulation, achieving 95.5% on LIBERO-Long and 65% on real-world chemistry tasks.
-
ERQA-Plus: A Diagnostic Benchmark for Reasoning in Embodied AI
ERQA-Plus is a new benchmark dataset with a structured taxonomy for evaluating fine-grained embodied reasoning in AI agents across perceptual, action-centric, social, navigation, and commonsense categories.
-
Embodied-R1.5: Evolving Physical Intelligence via Embodied Foundation Models
An 8B embodied foundation model trained on 15B tokens with multi-task RL and a Planner-Grounder-Corrector loop claims SOTA on 16/24 embodied VLM benchmarks and strong VLA/real-robot transfer.
-
Technical Report of RoboSpatial Challenge at CVPR 2026: Selective Reasoning Activation and Reference-Frame Disambiguation for Embodied Spatial Reasoning
RoboSpatialBrain applies forced <think> prefix activation and reference-frame redirection to RoboBrain2.5-8B-NV, achieving first place with 80.9% success on RoboSpatial-Home.