mlsanityData to 7 Oct 2026

Are socially-aware trajectory prediction models really socially-aware?

arXiv 2108.10879 ↗First posted 24 Aug 2021cs.CVcited by 32

Not on the map. Why?

Abstract (arXiv)

Our field has recently witnessed an arms race of neural network-based trajectory predictors. While these predictors are at the core of many applications such as autonomous navigation or pedestrian flow simulations, their adversarial robustness has not been carefully studied. In this paper, we introduce a socially-attended attack to assess the social understanding of prediction models in terms of collision avoidance. An attack is a small yet carefully-crafted perturbations to fail predictors. Technically, we define collision as a failure mode of the output, and propose hard- and soft-attention mechanisms to guide our attack. Thanks to our attack, we shed light on the limitations of the current models in terms of their social understanding. We demonstrate the strengths of our method on the recent trajectory prediction models. Finally, we show that our attack can be employed to increase the social understanding of state-of-the-art models. The code is available online: https://s-attack.github.io/

Key facts. Cited by 32 papers in the map (data through 7 Oct 2026). Builds on 29 of its 49 references that are in the map. First posted on arXiv 24 Aug 2021.

Builds on

29 of 49 references.

References · 29

  1. 2014Explaining and Harnessing Adversarial Examplescited by 7,287
  2. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  3. 2013Intriguing properties of neural networkscited by 5,114
  4. 2015DeepFool: a simple and accurate method to fool deep neural networkscited by 1,315
  5. 2018Robustness May Be at Odds with Accuracycited by 873
  6. 2018Social GAN: Socially Acceptable Trajectories with Generative Adversarial Networkscited by 687
  7. 2017One pixel attack for fooling deep neural networkscited by 579
  8. 2020Social-STGCNN: A Social Spatio-Temporal Graph Convolutional Neural Network for Human Trajectory Predictioncited by 324
  9. 2018SoPhie: An Attentive GAN for Predicting Paths Compliant to Social and Physical Constraintscited by 296
  10. 2021End-to-end Interpretable Neural Motion Plannercited by 240
  11. 2020It Is Not the Journey but the Destination: Endpoint Conditioned Trajectory Predictioncited by 235
  12. 2019Social-BiGAT: Multimodal Trajectory Forecasting using Bicycle-GAN and Graph Attention Networkscited by 228
17 more
  1. 2020Spatio-Temporal Graph Transformer Networks for Pedestrian Trajectory Predictioncited by 218
  2. 2017Social Attention: Modeling Attention in Human Crowdscited by 185
  3. 2020Human Trajectory Forecasting in Crowds: A Deep Learning Perspectivecited by 128
  4. 2016Decentralized Non-communicating Multiagent Collision Avoidance with Deep Reinforcement Learningcited by 104
  5. 2019Social Ways: Learning Multi-Modal Distributions of Pedestrian Trajectories with GANscited by 78
  6. 2019Adversarial Attacks on Deep Neural Networks for Time Series Classificationcited by 53
  7. 2019Adversarial Attacks on Time Seriescited by 40
  8. 2019Imperceptible Adversarial Attacks on Tabular Datacited by 35
  9. 2019Adversarial Attack on Skeleton-based Human Action Recognitioncited by 32
  10. 2017Context-Aware Trajectory Predictioncited by 31
  11. 2020Hold me tight! Influence of discriminative features on deep network boundariescited by 31
  12. 2021Injecting Knowledge in Data-driven Vehicle Trajectory Predictorscited by 18
  13. 2018An Evaluation of Trajectory Prediction Approaches and Notes on the TrajNet Benchmarkcited by 16
  14. 2020Pedestrian Intention Prediction: A Multi-task Perspectivecited by 16
  15. 2020Social-WaGDAT: Interaction-aware Trajectory Prediction via Wasserstein Graph Double-Attention Networkcited by 14
  16. 2019SafeCritic: Collision-Aware Trajectory Predictioncited by 3
  17. 2020MonStereo: When Monocular and Stereo Meet at the Tail of 3D Human Localizationcited by 1

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 3; 2022: 7; 2023: 11; 2024: 5; 2025: 4; 2026: 2.

32 papers.

Sort:

By area:

  1. 2022On Adversarial Robustness of Trajectory Prediction for Autonomous Vehiclescited by 89
  2. 2024UniTraj: A Unified Framework for Scalable Vehicle Trajectory Predictioncited by 51
  3. 2023Traj-MAE: Masked Autoencoders for Trajectory Predictioncited by 42
  4. 2021Vehicle trajectory prediction works, but not everywherecited by 39
  5. 2022A generic diffusion-based approach for 3D human pose prediction in the wildcited by 33
  6. 2023Social-Transmotion: Promptable Human Trajectory Predictioncited by 33
  7. 2024V2XPnP: Vehicle-to-Everything Spatio-Temporal Fusion for Multi-Agent Perception and Predictioncited by 31
  8. 2022Deep Interactive Motion Prediction and Planning: Playing Games with Motion Prediction Modelscited by 27
  9. 2022CausalAgents: A Robustness Benchmark for Motion Forecasting using Causal Relationshipscited by 18
  10. 2023Joint Metrics Matter: A Better Standard for Trajectory Forecastingcited by 17
  11. 2022Safety-compliant Generative Adversarial Networks for Human Trajectory Forecastingcited by 12
  12. 2022Motion Style Transfer: Modular Low-Rank Adaptation for Deep Motion Forecastingcited by 12
20 more
  1. 2022Pedestrian 3D Bounding Box Predictioncited by 11
  2. 2023Toward Reliable Human Pose Forecasting with Uncertaintycited by 11
  3. 2024Certified Human Trajectory Predictioncited by 6
  4. 2025Social-Pose: Enhancing Trajectory Prediction with Human Body Posecited by 5
  5. 2023JRDB-Traj: A Dataset and Benchmark for Trajectory Forecasting in Crowdscited by 4
  6. 2023Sim-to-Real Causal Transfer: A Metric Learning Approach to Causally-Aware Interaction Representationscited by 4
  7. 2024Testing autonomous vehicles and AI: perspectives and challenges from cybersecurity, transparency, robustness and fairnesscited by 3
  8. 2025Realistic Adversarial Attacks for Robustness Evaluation of Trajectory Prediction Models via Future State Perturbationcited by 3
  9. 2021SVG-Net: An SVG-based Trajectory Prediction Modelcited by 2
  10. 2021A Shared Representation for Photorealistic Driving Simulatorscited by 2
  11. 2025INTENT: Trajectory Prediction Framework with Intention-Guided Contrastive Clusteringcited by 2
  12. 2023Consistent Valid Physically-Realizable Adversarial Attack against Crowd-flow Prediction Modelscited by 1
  13. 2023Robustness Benchmark of Road User Trajectory Prediction Models for Automated Drivingcited by 1
  14. 2023Improving Trajectory Prediction in Dynamic Multi-Agent Environment by Dropping Waypointscited by 1
  15. 2023Enhancing Trajectory Prediction through Self-Supervised Waypoint Noise Predictioncited by 1
  16. 2024iMotion-LLM: Instruction-Conditioned Trajectory Generationcited by 1
  17. 2025HHI-Assist: A Dataset and Benchmark of Human-Human Interaction in Physical Assistance Scenariocited by 1
  18. 2023Semantic and Articulated Pedestrian Sensing Onboard a Moving Vehiclecited by 0
  19. 2026Super Agents and Confounders: Influence of surrounding agents on vehicle trajectory predictioncited by 0
  20. 2026TSCA-Net: Temporal-Spatial Clique Attention for Interpretable Multimodal Pedestrian Trajectory Predictioncited by 0