Description
Embodied navigation — the ability of autonomous agents to move purposefully through complex, unstructured environments — sits at the intersection of perception, cognition, and action. Despite significant progress driven by large-scale pretraining, state-of-the-art navigation systems still struggle with tasks that humans find straightforward: interpreting multi-step spatial instructions, reasoning about occluded regions, generalizing geometric priors across scene types, and composing spatial relations hierarchically. These gaps point to a fundamental challenge: current systems lack robust spatial intelligence — the capacity to form, manipulate, and exploit structured internal representations of the world for goal-directed locomotion.
SISREN brings together researchers from robot learning, computer vision, cognitive science, and natural language processing to examine where spatial intelligence breaks down in embodied agents and how structured reasoning — for instance over maps, scene graphs, affordances, and language-grounded representations — can address these failures. The event is aimed at researchers and graduate students working on embodied navigation, scene understanding, and representation learning, and is especially relevant to the CoRL community's focus on deployable, generalizable robotic systems.
Core Challenges / Research Questions
SISREN will focus on the following concrete, open research challenges:
- C1 Spatial representation learning. What inductive biases, architectures, or training objectives lead to representations that support downstream spatial reasoning, rather than merely fitting perception benchmarks?
- C2 Compositional and long-horizon spatial reasoning. How can agents decompose complex spatial instructions (“go past the chair, turn left at the hallway, stop in front of the blue door”) into executable sub-goals, and how should the knowledge of a sub-task’s failure propagate to subsequent tasks?
- C3 Generalizable cognitive maps. Can learned map representations transfer across environments with different geometry, semantic content, and sensor modalities? What is the right abstraction level (metric, topological, or semantic) for robust generalization?
- C4 Grounding language to spatial reasoning. How should language models interact with spatial representations? When is tight coupling beneficial; when does it introduce brittleness?
- C5 Uncertainty-aware navigation. How should agents represent and communicate uncertainty on spatial states, and how does structured reasoning improve decision-making under partial observability?
- C6 Evaluation methodology. Current benchmarks reward pattern matching rather than genuine spatial understanding. What evaluation protocols better distinguish structured reasoning from shortcut learning?
Schedule
Afternoon workshop · November 12th, 2026 · 2:00–6:00 PM · Room 306, JW Marriott Austin
| Time | Activity |
|---|---|
| 2:00–2:05 PM |
Opening remarks5 min
Workshop vision and overview of challenges C1 – C6
|
| 2:05–3:05 PM |
Invited talk block I60 min · 25 min talk + 5 min Q&A each
Dinesh Manocha — “Beyond Pixels and Prompts: Scaling Physical, Social, and Adversarial Safety in Vision-Language-Action Navigation”
C5C6
Bolei Zhou — “Scaling Long-Horizon Sidewalk Navigation: From Fleet Data and Human Interventions to World Models”
C3C5C6
|
| 3:05–3:35 PM |
Contributed spotlight session30 min · 5 min each
|
| 3:35–4:00 PM |
Afternoon coffee break & poster session25 min
Refreshments, poster presentations, live demos, and informal discussion.
|
| 4:00–4:20 PM |
Breakout discussion20 min
Participants self-select into challenge groups — A: C1 · B: C2C3 · C: C4 · D: C5C6. Each group produces a short summary for the panel.
|
| 4:20–5:20 PM |
Invited talk block II60 min · 25 min talk + 5 min Q&A each
Lu Gan — “Open-Vocabulary, Embodiment-Aware Maps for Navigation in the Wild”
C1C3
Jason Liu — “Robotic Language Grounding through Spatiotemporal Reasoning”
C2C4
|
| 5:20–5:50 PM |
Panel discussion & open Q&A30 min
Moderated discussion with all invited speakers; synthesis of breakout findings across C1 – C6
|
| 5:50–6:00 PM |
Awards and closing10 min
Awards, wrap-up, networking, and identification of key open problems.
|
Invited Speakers
All speakers confirmed.
Assistant Professor of Aerospace Engineering
Georgia Institute of Technology
Georgia Institute of Technology
“Open-Vocabulary, Embodiment-Aware Maps for Navigation in the Wild”
C1C3
Postdoctoral Fellow, Computer Science
MIT
MIT
“Robotic Language Grounding through Spatiotemporal Reasoning”
C2C4
Distinguished Professor of Computer Science
University of Maryland
University of Maryland
“Beyond Pixels and Prompts: Scaling Physical, Social, and Adversarial Safety in Vision-Language-Action Navigation”
C5C6
Associate Professor of Computer Science
UCLA
UCLA
“Scaling Long-Horizon Sidewalk Navigation: From Fleet Data and Human Interventions to World Models”
C3C5C6
Accepted Papers
- Iterative Reasoning About Clues that Matter for Navigation
- If, Then, Otherwise: Diagnosing Conditional Branching in Vision-Language Navigation
- Beyond Individual Intelligence: Learning Adaptive Multi-Robot Navigation
- VLD: Visual Learned Distance as an Interface for Reinforcement Learning Navigation
- Superpixels as Terrain Tokens: An Oversegmented Representation for Foothold Planning and Learned Tracking Control
- On the Surprising Effectiveness of Curriculum RL for Social Navigation with Sparse Human Localization
- Knowing Where to Go: Generalizing Mobile Manipulation to Unseen Environments from a Single Video
- DrivingBench: Can Vision-Language Models Drive a Toyota Corolla?
- Anytime Spatio-Semantic 3D Grounding for Online Navigation
- WayTrace: A Map-to-Language-to-Trace Hierarchy for Embodied Urban Navigation
- The Constitutional Controller: Doubt-Calibrated Steering of Compliant Agents
- RangeNav: Task-Aligned LiDAR Range-Image Pretraining for Robot Navigation
- CoopUQ: Robust Multi-Robot Social Navigation via Uncertainty-Aware Cooperative Forecasting
- Surfing the Wavelets: Multiresolution Online Feature Mapping for Scalable Open-Vocabulary 3D Perception
- The Surprising Spatial Awareness of a Frontier VLM in Drone Piloting Tasks
- Trajectory Prediction Meets Large Language Models: A Survey
- Hierarchical Implicit Grids for Geometry-Aware Language Grounding
- Does Navigation Fine-Tuning Increase Ego-Motion Decodability in VLA Hidden States?
- MobileVISTA: Generative Data Augmentation for Pose Generalization in Mobile Manipulation
- TrackletNav: Real-world Topological Navigation with Long-Term Capability in Complex Outdoor Scenes
- HINT-VLM: Humanoid Interactive Navigation over Rough Terrain with Embodiment-Grounded Vision Language Model
- MessyMem: Learning-from-Doing Memory for Mobile Manipulation
- MoMa-NMP: Scene-Scale Neural Motion Planning for Mobile Manipulators
- RAVEN: Long-Horizon Reasoning and Navigation with a Visuo-Spatial-Temporal Memory
- RepV: Scalable Neurosymbolic Plan Verification via Latent Space Alignment
- From Imitation to Alignment: Human-Preference Flow Policies for Long-Horizon Sidewalk Navigation
- Do Embodied VLMs Actually Use Their Scene-Graph Memory?
- WeaveRL: Weaving Reconstruction into Scene-Aware Fabrics for Perceptive Reinforcement Learning
- EliGSiR: Continual RGB-D Mapping with Gaussian Splatting under Bounded Compute
- NavWorld: Scaling Long-Horizon Outdoor Language Navigation with Crowdsourced Ground-Level Imagery
- OREN-X: Octree Residual Network for Real-Time Multi-Modal Mapping
- FrodoBots-27K: A Multi-Sensor Robot Navigation Dataset with Critical-Event Detection and Multi-Traverse Alignment
- WildOS: Open-Vocabulary Object Search in the Wild
- Depth Representation for Navigation and Collision Avoidance in Cluttered Environments
- The Constitutional Filter: Bayesian Estimation of Compliant Agents
Logistics
- Venue Room 306, JW Marriott Austin
- Posters We will have 29 posters. Authors of spotlight talks do not need to present a poster.
- Boards Each double-sided board will accommodate four posters (two on each side).
- Poster size Posters must be smaller than 4 ft × 4 ft (e.g., 48 in × 36 in) and sized so that two can fit side by side on each face.
Call for Papers
Submissions are now closed. Contributed papers were invited on topics mapped to the six challenge areas above and are non-archival to allow dual submission with venue proceedings.
Important Dates
Submission deadline
Oct 20, 2026
11:59 pm AoE · closed
Notification of results
Oct 25, 2026
Workshop
Nov 12, 2026
Room 306, JW Marriott Austin
Topics of Interest
Topics of interest include, but are not limited to:
- Spatial representation learningInductive biases, architectures, and training objectives for structured spatial understanding
- Compositional & long-horizon reasoningDecomposition of complex instructions, sub-goal planning, failure propagation
- Generalizable cognitive mapsMetric, topological, and semantic map representations; cross-environment transfer
- Language groundingInteraction between language models and spatial representations; tight vs. loose coupling tradeoffs
- Uncertainty-aware navigationUncertainty representation, communication, and decision-making under partial observability
- Evaluation methodologyBenchmarks and protocols that distinguish structured reasoning from shortcut learning
Submission Guidelines
Length
2–4 page extended abstract or up to 8-page full paper (plus references). These page limits are suggestions rather than strict requirements.
Format
Submissions may use any paper template and must be anonymized for double-blind review. We recommend either the CoRL 2026 LaTeX template or the IEEE conference template.
Reviews
Each submission will receive at least two double-blind reviews from a program committee spanning robotics, computer vision, NLP, and cognitive science. At least three spotlight slots are reserved for junior researchers (graduate students or postdocs as first authors).
Presentation
Accepted submissions are invited for a poster presentation at the workshop. Six papers were also selected for a contributed spotlight talk; spotlight presenters do not need to present a poster.
Dual submission
Published work, and work submitted concurrently to other conferences or workshops (including CoRL and IROS), is welcome.
Non-archival
Accepted submissions will be made available on the workshop website as non-archival reports, allowing subsequent submission to conferences and journals.
OpenReview
Submissions were handled through OpenReview. If you have any questions, do not hesitate to reach out to neel.bhatt@utdallas.edu.
Organizers
Neel P. Bhatt
University of Texas at Dallas
Dhruv Shah
Princeton University
Chen Wang
University at Buffalo