Pith. sign in

REVIEW 2 major objections 2 minor 1 cited by

Current embodied AI models cannot reliably follow social norms without explicit instructions.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

T0 review · grok-4.3

2026-06-30 23:25 UTC pith:HHHFOMJR

load-bearing objection RobotEQ introduces a new benchmark for unguided social-norm adherence in embodied AI, but the manual annotations lack any reported validation, so the performance gaps are hard to interpret. the 2 major comments →

arxiv 2605.06234 v2 pith:HHHFOMJR submitted 2026-05-07 cs.RO cs.HC

RobotEQ: Transitioning from Passive Intelligence to Active Intelligence in Embodied AI

Kuofei Fang , Xinyi Che , Haomin Ouyang , Shufan Zhang , Xuehao Wang , Qi Liu , Liyi Liu , Chenqi Zhang
show 7 more authors
Wenxi Cai Wenyu Dai Jinyang Wu Fan Zhang Haoyu Chen Bin He Zheng Lian
This is my paper
classification cs.RO cs.HC
keywords embodied AIactive intelligencesocial normsbenchmarkspatial groundingaction judgmentRAGrobotics
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper distinguishes user-guided passive intelligence from unguided active intelligence and introduces the RobotEQ benchmark to test whether models can judge permissible robot actions from egocentric images alone. It builds a dataset of 1,894 images across 10 categories and 56 subcategories, annotated with 4,944 action-judgment questions and 1,157 spatial-grounding questions that encode social norms. Experiments show that state-of-the-art models perform poorly overall and especially on spatial grounding, while retrieval-augmented generation from external social-norm knowledge bases improves results. The work positions this benchmark as a step toward robots that comply with societal rules without constant human direction.

Core claim

RobotEQ is the first benchmark for active intelligence in embodied AI; it consists of RobotEQ-Data (1,894 egocentric images with 4,944 action-judgment and 1,157 spatial-grounding questions across 10 categories) and RobotEQ-Bench, which demonstrates that current models fall short in reliable active intelligence, particularly spatial grounding, while RAG with external social-norm knowledge generally improves performance.

What carries the argument

RobotEQ-Data and RobotEQ-Bench, which supply manually annotated action-judgment and spatial-grounding questions that specify appropriate robot actions from egocentric images according to social norms.

Load-bearing premise

The 4,944 action-judgment and 1,157 spatial-grounding questions produced by manual annotation faithfully represent the social norms that should govern robot behavior across the 10 categories and 56 subcategories.

What would settle it

A new set of real-world human judgments on the same images that systematically disagrees with the paper's manual annotations on which actions are permissible.

Watch this falsifier — get emailed when new claim-graph text bears on it.

If this is right

  • Models must improve spatial grounding to achieve reliable active intelligence.
  • Adding retrieval from external social-norm knowledge bases raises performance on action judgments.
  • Robotics can move from passive, instruction-following manipulation to active social compliance.
  • Systematic evaluation across 10 categories reveals consistent gaps in current approaches.
  • The benchmark supplies a concrete testbed for measuring progress toward unguided norm adherence.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Robots passing this benchmark could operate with less direct supervision in shared human spaces.
  • The dataset could be extended with video or multi-view inputs to test dynamic norm understanding.
  • Cultural variation in norms would require separate annotation efforts or modular knowledge bases.
  • Combining RobotEQ scores with existing task-completion benchmarks could measure overall social competence.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

2 major / 2 minor

Summary. The manuscript introduces RobotEQ as the first benchmark for active intelligence in embodied AI, where robots must infer and adhere to social norms without explicit user instructions (contrasted with passive intelligence). It describes construction of RobotEQ-Data from 1,894 egocentric images across 10 categories and 56 subcategories, yielding 4,944 action-judgment questions and 1,157 spatial-grounding questions via manual annotation. RobotEQ-Bench then evaluates state-of-the-art models, reporting that they fall short especially on spatial grounding while RAG with external social-norm knowledge bases improves results.

Significance. If the benchmark is shown to validly encode the relevant norms, the work would usefully quantify a gap between current models' passive task-following and the active social compliance needed for real-world robot deployment, while providing an initial testbed and a simple RAG mitigation that could inform follow-on research in socially aware robotics.

major comments (2)
  1. [RobotEQ-Data] RobotEQ-Data section: the manual annotation process that produced the 4,944 action-judgment and 1,157 spatial-grounding questions reports neither inter-annotator agreement statistics nor an expert-review protocol, nor any handling of cultural or contextual variability; because the headline claims about model shortfalls rest directly on these questions faithfully representing the norms that should govern robot behavior, the absence of validation metrics leaves the measured performance gaps difficult to interpret.
  2. [Experiments] Experiments / RobotEQ-Bench section: the reported performance gaps (especially the spatial-grounding shortfall) are presented without statistical significance tests or confidence intervals, so it is unclear whether the observed differences between models and between RAG vs. non-RAG conditions exceed what would be expected from sampling variability alone.
minor comments (2)
  1. [Abstract] Abstract: the definitions of 'passive intelligence' and 'active intelligence' are introduced only by example; a concise formal distinction would help readers map the new terminology onto existing embodied-AI literature.
  2. [RobotEQ-Data] The paper does not discuss how the 10 categories and 56 subcategories were selected or whether they were chosen to be representative of broader social-norm distributions.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive comments on our manuscript. We address each major comment point by point below, indicating planned revisions where appropriate.

read point-by-point responses

  1. Referee: [RobotEQ-Data] RobotEQ-Data section: the manual annotation process that produced the 4,944 action-judgment and 1,157 spatial-grounding questions reports neither inter-annotator agreement statistics nor an expert-review protocol, nor any handling of cultural or contextual variability; because the headline claims about model shortfalls rest directly on these questions faithfully representing the norms that should govern robot behavior, the absence of validation metrics leaves the measured performance gaps difficult to interpret.

    Authors: We agree that explicit reporting of inter-annotator agreement, expert review, and cultural handling would improve interpretability. In the revision we will expand the RobotEQ-Data section with these details: we computed Fleiss' kappa on a 20% overlap subset (value to be reported), included a two-stage expert review by robotics and ethics researchers, and addressed cultural variability by recruiting annotators from multiple regions with consensus resolution for disagreements. These additions will directly support the benchmark's validity claims. revision: yes

  2. Referee: [Experiments] Experiments / RobotEQ-Bench section: the reported performance gaps (especially the spatial-grounding shortfall) are presented without statistical significance tests or confidence intervals, so it is unclear whether the observed differences between models and between RAG vs. non-RAG conditions exceed what would be expected from sampling variability alone.

    Authors: We concur that statistical rigor is needed. The revision will add bootstrap-derived 95% confidence intervals for all accuracy metrics and paired significance tests (McNemar's test for model comparisons and Wilcoxon signed-rank for RAG vs. baseline) to confirm that reported gaps exceed sampling variability. revision: yes

Circularity Check

0 steps flagged

No circularity in benchmark construction or model evaluation

full rationale

The paper's chain consists of manual annotation to produce RobotEQ-Data (4,944 action-judgment and 1,157 spatial-grounding questions from 1,894 images) followed by direct empirical evaluation of models on the held-out RobotEQ-Bench. Reported shortfalls in active intelligence and RAG gains are measured performance numbers, not algebraic reductions, fitted-parameter renamings, or self-referential derivations. No equations, self-citations for uniqueness theorems, or ansatzes appear in the provided methodology. The benchmark is self-contained against external model testing.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the assumption that the manually chosen categories and annotations capture the relevant social norms; no free parameters are fitted to performance data and no new physical entities are postulated.

free parameters (1)
  • Category and subcategory selection
    10 representative embodied categories and 56 subcategories chosen by the authors to span typical robot scenarios.
axioms (1)
  • domain assumption Social norms relevant to robot behavior can be adequately captured by binary action-judgment questions and spatial-grounding questions on static egocentric images.
    Invoked when constructing the 4,944 + 1,157 questions from the 1,894 images.

pith-pipeline@v0.9.1-grok · 5801 in / 1256 out tokens · 19081 ms · 2026-06-30T23:25:44.346640+00:00 · methodology

0 comments
read the original abstract

Embodied AI is a prominent research topic in both academia and industry. Current research centers on completing tasks based on explicit user instructions. However, for robots to integrate into human society, they must understand which actions are permissible and which are prohibited, even without explicit commands. We refer to the user-guided AI as passive intelligence and the unguided AI as active intelligence. This paper introduces RobotEQ, the first benchmark for active intelligence, aiming to assess whether existing models can comprehend and adhere to social norms in embodied scenarios. First, we construct RobotEQ-Data, a dataset consisting of 1,894 egocentric images, spanning 10 representative embodied categories and 56 subcategories. Through extensive manual annotation, we provide 4,944 action judgment questions and 1,157 spatial grounding questions, specifying appropriate robot actions across diverse scenarios. Furthermore, we establish RobotEQ-Bench to evaluate the performance of state-of-the-art models on this task. Experimental results demonstrate that current models still fall short in achieving reliable active intelligence, particularly in spatial grounding. Meanwhile, leveraging RAG techniques to incorporate external social norm knowledge bases can generally enhance performance. This work can facilitate the transition of robotics from user-guided passive manipulation to active social compliance.

Figures

Figures reproduced from arXiv: 2605.06234 by Bin He, Chenqi Zhang, Fan Zhang, Haomin Ouyang, Haoyu Chen, Jinyang Wu, Kuofei Fang, Liyi Liu, Qi Liu, Shufan Zhang, Wenxi Cai, Wenyu Dai, Xinyi Che, Xuehao Wang, Zheng Lian.

Figure 1
Figure 1. Figure 1: RobotEQ. This benchmark consists of multiple robot-view images covering typical embodied categories and subcategories. It provides two types of questions: action judgment and spatial grounding. For action judgment, both proper and improper actions are annotated; for spatial grounding, both appropriate and inappropriate regions or movement trajectories are labeled. whether robots can successfully complete t… view at source ↗
Figure 2
Figure 2. Figure 2: Data collection pipeline. 1) Scenario design. We define scenario categories and subcate￾gories, and then employ LLMs to generate diverse image descriptions. 2) Image generation. These descriptions serve as input for image generation. Since generated images may contain artifacts, we further refine them using image editing. 3) Action judgment. For each image, we compile a list of candidate actions and annota… view at source ↗
Figure 3
Figure 3. Figure 3: Overview of RobotEQ-Data. (a) Key statistics of the benchmark. (b) Distribution of the ten scenario categories. (c) Distribution of the eight evaluation dimensions. which was subsequently calibrated by a domain expert to establish the final ground truth. Based on annotator accuracy, we selected the 7 highest-performing annotators to form the formal labeling team. This pilot phase ensured the reliability an… view at source ↗
Figure 4
Figure 4. Figure 4: Dimension-level action judgment performance. Radar charts compare representative models with human performance across the eight dimensions in RobotEQ-Bench. Qwen3-VL-8B Claude Sonnet 4.6 GUI-Actor-7B Gemma-3-12B Claude Opus 4.7 GPT-5.5 Nanonets-OCR2 GroundNext-7B Gemini 2.5 Pro InfiGUI-G1 InternVL3-8B LLaVA-OneVision GLM-4.1V-9B DeepSeek-VL2 45.0 47.5 50.0 52.5 55.0 57.5 60.0 62.5 Macro-F1 (%) Macro-F1 (%)… view at source ↗
Figure 5
Figure 5. Figure 5: Spatial grounding. Human performance is annotated alongside each subplot title. 5.3 Error Analysis To better understand model limitations, we examine representative GPT-5.5 [33] errors on action judgment and spatial grounding in view at source ↗
Figure 6
Figure 6. Figure 6: Representative error cases from GPT-5.5. We categorize failures into four types: Overly Aggressive, Overly Cautious, Lack of Social Experience, and Spatial Grounding Error. For CoT prompting, we guide the model to reason through a fixed sequence before making the final judgment: scene analysis, demand recognition, role reflection, and action assessment. This prompt encourages the model to consider both the… view at source ↗
Figure 7
Figure 7. Figure 7: presents the complete taxonomy. We briefly summarize the 10 major categories below view at source ↗
Figure 8
Figure 8. Figure 8: Prompt templates for scenario generation. Overview of the beam-phase and merge-phase prompts used in RobotEQ-Data, highlighting the input fields, generation constraints, deduplication rules, and expected output structure view at source ↗
Figure 9
Figure 9. Figure 9: Representative scenario examples. Five example scenarios illustrating how embodied agents must reason over nonverbal cues, spatial relations, and context-specific social norms in real￾world human environments. B Image Generation The scenarios produced by the beginning of the generation pipeline in Section 3.1 are textual descriptions. They specify the social context, the position of the agent, and the envi… view at source ↗
Figure 10
Figure 10. Figure 10: Scenario-to-image prompt synthesis. An example of how RobotEQ-Data converts a structured embodied social scenario into a visual prompt for image generation. The prompt preserves the social interaction conflict, specifies visual anchors and spatial relations, and produces a first￾person scene image for benchmark construction. Image Generation. The synthesized visual prompts are then used to generate candid… view at source ↗
Figure 11
Figure 11. Figure 11: Examples of image refinement. Representative raw and edited images from the automated refinement stage. The examples illustrate how the editing process improves visual grounding and scenario fidelity while preserving the intended embodied social context. Human Verification. After the automated revision stage, we aggregate the original image, the edited image, the corresponding scenario, and the scenario d… view at source ↗
Figure 12
Figure 12. Figure 12: Examples of the Label Studio annotation interface. The left panel shows the human verification stage where annotators compare original and edited scenario images, and the right panel shows the human annotation stage for action judgment and spatial grounding labelling. Additional cases are omitted for brevity. C Action Generation The action generation stage aims to construct, for each validated scenario, a… view at source ↗
Figure 13
Figure 13. Figure 13: Action generation prompt. Illustration of the prompt structure used to generate candidate action pools from a scenario image and its textual description. the benchmark, such as physically impossible actions, irrelevant actions, or actions that do not form a meaningful test of active intelligence. All annotations are collected through a Label Studio interface configured for this task. Pilot Study. To calib… view at source ↗
Figure 14
Figure 14. Figure 14: Action judgment evaluation example. The figure illustrates the input format used for action judgment in RobotEQ. Given a first-person scenario image, the model receives a role-specific question and a list of candidate actions, and must assign each action a binary label indicating whether it should or should not be performed. Please select the signal area in the diagram that you believe indicates a custome… view at source ↗
Figure 15
Figure 15. Figure 15: Comparison of sptaial grounding question generation pipelines. Representative examples comparing the two-stage and one-stage construction procedures for spatial grounding questions. The two-stage pipeline produces more precise and visually grounded spatial annotations, while the one-stage pipeline is more prone to misplaced, overly broad, or spatially incoherent annotations. produces generic questions or … view at source ↗
Figure 16
Figure 16. Figure 16: Spatial grounding evaluation example. The figure illustrates the input and output format for a spatially grounded multiple-choice question in RobotEQ-Data. Given an annotated robot-view scene image and a question, the model selects all applicable spatial regions and provides a brief rationale for its prediction. selected in Appendix D label spatial grounding questions through a Label Studio interface. We … view at source ↗
Figure 17
Figure 17. Figure 17: Chain-of-Thought prompt design for action judgment. The figure illustrates the CoT input view at source ↗
Figure 18
Figure 18. Figure 18: Example of a role-specific RAG knowledge base. The figure shows a representative view at source ↗

Review history (2 revisions) →

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. OPID: On-Policy Skill Distillation for Agentic Reinforcement Learning

    cs.CL 2026-06 unverdicted novelty 4.0

    OPID distills episode- and step-level skills from completed on-policy trajectories, routes them via critical-first mechanism, and combines the resulting log-probability shift advantage with outcome advantage for polic...

Reference graph

Works this paper leans on

53 extracted references · 7 canonical work pages · cited by 1 Pith paper · 6 internal anchors

  1. [1]

    Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs

    Abdelrahman Abouelenin, Atabak Ashfaq, Adam Atkinson, Hany Awadalla, Nguyen Bach, Jianmin Bao, Alon Benhaim, Martin Cai, Vishrav Chaudhary, Congcong Chen, et al. Phi-4-mini technical report: Compact yet powerful multimodal language models via mixture-of-loras. arXiv preprint arXiv:2503.01743, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  2. [2]

    Pravesh Agrawal, Szymon Antoniak, Emma Bou Hanna, Baptiste Bout, Devendra Chaplot, Jessica Chudnovsky, Diogo Costa, Baudouin De Monicault, Saurabh Garg, Theophile Gervet, Soham Ghosh, Amélie Héliou, Paul Jacob, Albert Q. Jiang, Kartik Khandelwal, Timothée Lacroix, Guillaume Lample, Diego Las Casas, Thibaut Lavril, Teven Le Scao, Andy Lo, William Marshall,...

    2024

  3. [3]

    Reid, Stephen Gould, and Anton van den Hengel

    Peter Anderson, Qi Wu, Damien Teney, Jake Bruce, Mark Johnson, Niko Sünderhauf, Ian D. Reid, Stephen Gould, and Anton van den Hengel. Vision-and-language navigation: Interpreting visually-grounded navigation instructions in real environments. In2018 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2018, Salt Lake City, UT, USA, June 18-22,...

    2018

  4. [4]

    System Card: Claude Opus 4.6

    Anthropic. System Card: Claude Opus 4.6. https://www-cdn.anthropic.com/0dd8650 75ad3132672ee0ab40b05a53f14cf5288.pdf, February 2026. Released February 5, 2026. 212 pages. Also available athttps://www.anthropic.com/system-cards

    2026

  5. [5]

    System Card: Claude Opus 4.7

    Anthropic. System Card: Claude Opus 4.7. https://www.anthropic.com/system-cards, April 2026. Released April 16, 2026. 232 pages. Download PDF from the System Cards page

    2026

  6. [6]

    System Card: Claude Sonnet 4.6

    Anthropic. System Card: Claude Sonnet 4.6. https://www-cdn.anthropic.com/78073 f739564e986ff3e28522761a7a0b4484f84.pdf, February 2026. Released February 2026. Also available athttps://www.anthropic.com/system-cards

    2026

  7. [7]

    Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond, 2023

    Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond, 2023

    2023

  8. [8]

    Qwen3-VL Technical Report

    Shuai Bai, Yuxuan Cai, Ruizhe Chen, Keqin Chen, Xionghui Chen, Zesen Cheng, Lianghao Deng, Wei Ding, Chang Gao, Chunjiang Ge, et al. Qwen3-vl technical report.arXiv preprint arXiv:2511.21631, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  9. [9]

    Qwen2.5-vl technical report, 2025

    Shuai Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Sibo Song, Kai Dang, Peng Wang, Shijie Wang, Jun Tang, Humen Zhong, Yuanzhi Zhu, Mingkun Yang, Zhaohai Li, Jianqiang Wan, Pengfei Wang, Wei Ding, Zheren Fu, Yiheng Xu, Jiabo Ye, Xi Zhang, Tianbao Xie, Zesen Cheng, Hang Zhang, Zhibo Yang, Haiyang Xu, and Junyang Lin. Qwen2.5-vl technical report, 2025

    2025

  10. [10]

    Seed 1.6 Technical Report

    ByteDance Seed Team. Seed 1.6 Technical Report. https://seed.bytedance.com/en/se ed1_6, 2025. Chinese version:https://research.doubao.com/zh/seed1_6. 10

    2025

  11. [11]

    Janus-pro: Unified multimodal understanding and generation with data and model scaling, 2025

    Xiaokang Chen, Zhiyu Wu, Xingchao Liu, Zizheng Pan, Wen Liu, Zhenda Xie, Xingkai Yu, and Chong Ruan. Janus-pro: Unified multimodal understanding and generation with data and model scaling, 2025

    2025

  12. [12]

    Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities

    Gheorghe Comanici, Eric Bieber, Mike Schaekermann, Ice Pasupat, Noveen Sachdeva, Inderjit Dhillon, Marcel Blistein, Ori Ram, Dan Zhang, Evan Rosen, et al. Gemini 2.5: Pushing the frontier with advanced reasoning, multimodality, long context, and next generation agentic capabilities.arXiv preprint arXiv:2507.06261, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  13. [13]

    Aya vision: Advancing the frontier of multilingual multimodality, 2025

    Saurabh Dash, Yiyang Nan, John Dang, Arash Ahmadian, Shivalika Singh, Madeline Smith, Bharat Venkitesh, Vlad Shmyhlo, Viraat Aryabumi, Walter Beller-Morales, Jeremy Pekmez, Jason Ozuzu, Pierre Richemond, Acyr Locatelli, Nick Frosst, Phil Blunsom, Aidan Gomez, Ivan Zhang, Marzieh Fadaee, Manoj Govindassamy, Sudip Roy, Matthias Gallé, Beyza Ermis, Ahmet Üst...

    2025

  14. [14]

    Danny Driess, Fei Xia, Mehdi S. M. Sajjadi, Corey Lynch, Aakanksha Chowdhery, Brian Ichter, Ayzaan Wahid, Jonathan Tompson, Quan Vuong, Tianhe Yu, Wenlong Huang, Yevgen Chebotar, Pierre Sermanet, Daniel Duckworth, Sergey Levine, Vincent Vanhoucke, Karol Hausman, Marc Toussaint, Klaus Greff, Andy Zeng, Igor Mordatch, and Pete Florence. Palm-e: an embodied ...

    2023

  15. [15]

    Rodriguez, Nicolas Chapados, David Vazquez, Adriana Romero-Soriano, Reihaneh Rabbany, Perouz Taslakian, Christopher Pal, Spandana Gella, and Sai Rajeswar

    Aarash Feizi, Shravan Nayak, Xiangru Jian, Kevin Qinghong Lin, Kaixin Li, Rabiul Awal, Xing Han Lù, Johan Obando-Ceron, Juan A. Rodriguez, Nicolas Chapados, David Vazquez, Adriana Romero-Soriano, Reihaneh Rabbany, Perouz Taslakian, Christopher Pal, Spandana Gella, and Sai Rajeswar. Grounding computer use agents on human demonstrations, 2025

    2025

  16. [16]

    Seedream 3.0 technical report, 2025

    Yu Gao, Lixue Gong, Qiushan Guo, Xiaoxia Hou, Zhichao Lai, Fanshi Li, Liang Li, Xiaochen Lian, Chao Liao, Liyang Liu, Wei Liu, Yichun Shi, Shiqi Sun, Yu Tian, Zhi Tian, Peng Wang, Rui Wang, Xuanda Wang, Xun Wang, Ye Wang, Guofeng Wu, Jie Wu, Xin Xia, Xuefeng Xiao, Zhonghua Zhai, Xinyu Zhang, Qi Zhang, Yuwei Zhang, Shijia Zhao, Jianchao Yang, and Weilin Hu...

    2025

  17. [17]

    Gemini 3 Pro Image Model Card

    Google DeepMind. Gemini 3 Pro Image Model Card. https://storage.googleapis.com /deepmind-media/Model-Cards/Gemini-3-Pro-Image-Model-Card.pdf , November

  18. [18]

    Released November 20, 2025

    2025

  19. [19]

    Gemini 3.1 Pro Model Card

    Google DeepMind. Gemini 3.1 Pro Model Card. https://deepmind.google/models/mod el-cards/gemini-3-1-pro/ , February 2026. PDF version: https://storage.google apis.com/deepmind-media/Model-Cards/Gemini-3-1-Pro-Model-Card.pdf

    2026

  20. [20]

    Navigating the digital world as humans do: Universal visual grounding for GUI agents

    Boyu Gou, Ruohan Wang, Boyuan Zheng, Yanan Xie, Cheng Chang, Yiheng Shu, Huan Sun, and Yu Su. Navigating the digital world as humans do: Universal visual grounding for GUI agents. InThe Thirteenth International Conference on Learning Representations, 2025

    2025

  21. [21]

    The Llama 3 Herd of Models

    Aaron Grattafiori, Abhimanyu Dubey, Abhinav Jauhri, Abhinav Pandey, Abhishek Kadian, Ahmad Al-Dahle, Aiesha Letman, Akhil Mathur, Alan Schelten, Alex Vaughan, et al. The llama 3 herd of models.arXiv preprint arXiv:2407.21783, 2024

    work page internal anchor Pith review Pith/arXiv arXiv 2024

  22. [22]

    GLM-4.5V and GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning

    Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, et al. Glm-4.5 v and glm-4.1 v-thinking: Towards versatile multimodal reasoning with scalable reinforcement learning.arXiv preprint arXiv:2507.01006, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  23. [23]

    Inner monologue: Em- bodied reasoning through planning with language models

    Wenlong Huang, Fei Xia, Ted Xiao, Harris Chan, Jacky Liang, Pete Florence, Andy Zeng, Jonathan Tompson, Igor Mordatch, Yevgen Chebotar, Pierre Sermanet, Tomas Jackson, Noah Brown, Linda Luu, Sergey Levine, Karol Hausman, and brian ichter. Inner monologue: Em- bodied reasoning through planning with language models. In Karen Liu, Dana Kulic, and Jeff Ichnow...

    2023

  24. [24]

    Joshi, Kyle Jeffrey, Rosario Jauregui Ruano, Jasmine Hsu, Keerthana Gopalakrishnan, Byron David, Andy Zeng, and Chuyuan Kelly Fu

    brian ichter, Anthony Brohan, Yevgen Chebotar, Chelsea Finn, Karol Hausman, Alexander Herzog, Daniel Ho, Julian Ibarz, Alex Irpan, Eric Jang, Ryan Julian, Dmitry Kalashnikov, Sergey Levine, Yao Lu, Carolina Parada, Kanishka Rao, Pierre Sermanet, Alexander T Toshev, Vincent Vanhoucke, Fei Xia, Ted Xiao, Peng Xu, Mengyuan Yan, Noah Brown, Michael Ahn, Omar ...

    2023

  25. [25]

    Building and better understanding vision-language models: insights and future directions, 2024

    Hugo Laurençon, Andrés Marafioti, Victor Sanh, and Léo Tronchon. Building and better understanding vision-language models: insights and future directions, 2024

    2024

  26. [26]

    LLaV A-onevision: Easy visual task transfer

    Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Peiyuan Zhang, Yanwei Li, Ziwei Liu, and Chunyuan Li. LLaV A-onevision: Easy visual task transfer. Transactions on Machine Learning Research, 2025

    2025

  27. [27]

    Aligning cyber space with physical world: A comprehensive survey on embodied ai, 2025

    Yang Liu, Weixing Chen, Yongjie Bai, Xiaodan Liang, Guanbin Li, Wen Gao, and Liang Lin. Aligning cyber space with physical world: A comprehensive survey on embodied ai, 2025

    2025

  28. [28]

    Infigui-g1: Advancing gui grounding with adaptive exploration policy optimization

    Yuhang Liu, Zeyu Liu, Shuanghe Zhu, Pengxiang Li, Congkai Xie, Jiasheng Wang, Xueyu Hu, Xiaotian Han, Jianbo Yuan, Xinyao Wang, et al. Infigui-g1: Advancing gui grounding with adaptive exploration policy optimization. InProceedings of the AAAI Conference on Artificial Intelligence, pages 32267–32275, 2026

    2026

  29. [29]

    Advancing social intelligence in ai agents: Technical challenges and open questions

    Leena Mathur, Paul Pu Liang, and Louis-Philippe Morency. Advancing social intelligence in ai agents: Technical challenges and open questions. InProceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 20541–20560, 2024

    2024

  30. [30]

    Human-level control through deep reinforcement learning.Nature, 518(7540):529–533, 2015

    V olodymyr Mnih, Koray Kavukcuoglu, David Silver, Andrei A Rusu, Joel Veness, Marc G Bellemare, Alex Graves, Martin Riedmiller, Andreas K Fidjeland, Georg Ostrovski, et al. Human-level control through deep reinforcement learning.Nature, 518(7540):529–533, 2015

    2015

  31. [31]

    Human behavior atlas: Benchmarking unified psychological and social behavior understanding.arXiv preprint arXiv:2510.04899, 2025

    Keane Ong, Wei Dai, Carol Li, Dewei Feng, Hengzhi Li, Jingyao Wu, Jiaee Cheong, Rui Mao, Gianmarco Mengaldo, Erik Cambria, et al. Human behavior atlas: Benchmarking unified psychological and social behavior understanding.arXiv preprint arXiv:2510.04899, 2025

    work page arXiv 2025

  32. [32]

    GPT-4o System Card, 2024

    OpenAI. GPT-4o System Card, 2024. Covers GPT-4o and GPT-4o-mini

    2024

  33. [33]

    GPT-5.4 Thinking System Card

    OpenAI. GPT-5.4 Thinking System Card. https://deploymentsafety.openai.com/gp t-5-4-thinking, March 2026. Released March 5, 2026

    2026

  34. [34]

    GPT-5.5 System Card

    OpenAI. GPT-5.5 System Card. https://deploymentsafety.openai.com/gpt-5-5 , April 2026. Released April 23, 2026

    2026

  35. [35]

    Manso, Anaís Garrell, Alberto Sanfeliu, Anne Spalanzani, and Rachid Alami

    Phani Teja Singamaneni, Pilar Bachiller-Burgos, Luis J. Manso, Anaís Garrell, Alberto Sanfeliu, Anne Spalanzani, and Rachid Alami. A survey on socially aware robot navigation: Taxonomy and future challenges.The International Journal of Robotics Research, 43(10):1533–1572, February 2024

    2024

  36. [36]

    Gui-g2: Gaussian reward modeling for gui grounding, 2025

    Fei Tang, Zhangxuan Gu, Zhengxi Lu, Xuyang Liu, Shuheng Shen, Changhua Meng, Wen Wang, Wenqi Zhang, Yongliang Shen, Weiming Lu, Jun Xiao, and Yueting Zhuang. Gui-g2: Gaussian reward modeling for gui grounding, 2025

    2025

  37. [37]

    Gemma 3 Technical Report

    Gemma Team. Gemma 3 technical report.arXiv preprint arXiv:2503.19786, 2025

    work page internal anchor Pith review Pith/arXiv arXiv 2025

  38. [38]

    GUI-actor: Coordinate-free visual grounding for GUI agents

    Qianhui Wu, Kanzhi Cheng, Rui Yang, Chaoyun Zhang, Jianwei Yang, Huiqiang Jiang, Jian Mu, Baolin Peng, Bo Qiao, Reuben Tan, Si Qin, Lars Liden, Qingwei Lin, Huan Zhang, Tong Zhang, Jianbing Zhang, Dongmei Zhang, and Jianfeng Gao. GUI-actor: Coordinate-free visual grounding for GUI agents. InThe Thirty-ninth Annual Conference on Neural Information Processi...

    2026

  39. [39]

    Deepseek-vl2: Mixture-of-experts vision-language models for advanced multimodal understanding, 2024

    Zhiyu Wu, Xiaokang Chen, Zizheng Pan, Xingchao Liu, Wen Liu, Damai Dai, Huazuo Gao, Yiyang Ma, Chengyue Wu, Bingxuan Wang, Zhenda Xie, Yu Wu, Kai Hu, Jiawei Wang, Yaofeng Sun, Yukun Li, Yishi Piao, Kang Guan, Aixin Liu, Xin Xie, Yuxiang You, Kai Dong, Xingkai Yu, Haowei Zhang, Liang Zhao, Yisong Wang, and Chong Ruan. Deepseek-vl2: Mixture-of-experts visio...

    2024

  40. [40]

    React: Synergizing reasoning and acting in language models

    Shunyu Yao, Jeffrey Zhao, Dian Yu, Nan Du, Izhak Shafran, Karthik R Narasimhan, and Yuan Cao. React: Synergizing reasoning and acting in language models. InThe Eleventh International Conference on Learning Representations, 2023

    2023

  41. [41]

    Social- iq: A question answering benchmark for artificial social intelligence

    Amir Zadeh, Michael Chan, Paul Pu Liang, Edmund Tong, and Louis-Philippe Morency. Social- iq: A question answering benchmark for artificial social intelligence. InProceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), June 2019

    2019

  42. [42]

    Tensor fusion network for multimodal sentiment analysis

    Amir Zadeh, Minghai Chen, Soujanya Poria, Erik Cambria, and Louis-Philippe Morency. Tensor fusion network for multimodal sentiment analysis. InProceedings of the 2017 conference on empirical methods in natural language processing, pages 1103–1114, 2017

    2017

  43. [43]

    Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph

    AmirAli Bagher Zadeh, Paul Pu Liang, Soujanya Poria, Erik Cambria, and Louis-Philippe Morency. Multimodal language analysis in the wild: Cmu-mosei dataset and interpretable dynamic fusion graph. InProceedings of the 56th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2236–2246, 2018

    2018

  44. [44]

    Internvl3: Exploring advanced training and test-time recipes for open-source multimodal models, 2025

    Jinguo Zhu, Weiyun Wang, Zhe Chen, Zhaoyang Liu, Shenglong Ye, Lixin Gu, Hao Tian, Yuchen Duan, Weijie Su, Jie Shao, Zhangwei Gao, Erfei Cui, Xuehui Wang, Yue Cao, Yangzhou Liu, Xingguang Wei, Hongjie Zhang, Haomin Wang, Weiye Xu, Hao Li, Jiahao Wang, Nianchen Deng, Songze Li, Yinan He, Tan Jiang, Jiapeng Luo, Yi Wang, Conghui He, Botian Shi, Xingcheng Zh...

    2025

  45. [45]

    Sanketi, Grecia Salazar, Michael S

    Brianna Zitkovich, Tianhe Yu, Sichun Xu, Peng Xu, Ted Xiao, Fei Xia, Jialin Wu, Paul Wohlhart, Stefan Welker, Ayzaan Wahid, Quan Vuong, Vincent Vanhoucke, Huong Tran, Radu Soricut, Anikait Singh, Jaspiar Singh, Pierre Sermanet, Pannag R. Sanketi, Grecia Salazar, Michael S. Ryoo, Krista Reymann, Kanishka Rao, Karl Pertsch, Igor Mordatch, Henryk Michalewski...

    2023

  46. [46]

    Non-verbal Signal Recognition: The ability to interpret non-verbal communicative cues, including gaze direction, hand gestures, body posture, head movements, pointing, beckoning, and other implicit signals such as chin-directed requests. 26

  47. [47]

    Proxemics & Spatial Norms: The ability to reason about personal space, appropriate pass- ing distance, queuing, yielding, spatial occlusion, positional relationships, and movement boundaries in shared environments

  48. [48]

    Role Boundary & Authority: The ability to recognize role-defined responsibilities and authority relations, including who may issue instructions, whether a request is legitimate, and whether an action oversteps age-, identity-, responsibility-, or organization-based boundaries

  49. [49]

    Timing & Interruption Norms: The ability to judge when to intervene, wait, interrupt, or yield, taking into account turn-taking conventions, ongoing interactions, sequential order, and the pacing of human activities

  50. [50]

    Contextual Volume & Behavioral Restraint: The ability to adjust voice volume, notifica- tion sounds, movement amplitude, and behavioral conspicuousness according to the social and environmental context

  51. [51]

    Resource & Ownership Norms: The ability to reason about ownership, borrowing, sharing, occupation rights, unattended belongings, and whether an object may be moved, used, returned, or left untouched

  52. [52]

    Priority & Protected Persons: The ability to identify people who require prioritized assistance or protection, such as children, elderly people, patients, vulnerable individuals, or people involved in emergency situations

  53. [53]

    Annotation methodology.We assign dimension labels through a two-stage process that combines LLM-based classification with human calibration

    Culture-Specific Norms: The ability to recognize etiquette, taboos, ceremonial practices, religious norms, and behavioral boundaries that vary across cultural or occasion-specific contexts. Annotation methodology.We assign dimension labels through a two-stage process that combines LLM-based classification with human calibration. In the first stage, Gemini...