Chain Of Interaction Benchmark (COIN): When Reasoning meets Embodied Interaction

• URL: http://arxiv.org/abs/2604.16886v1
• Date: Sat, 18 Apr 2026 07:26:24 GMT
• Title: Chain Of Interaction Benchmark (COIN): When Reasoning meets Embodied Interaction
• Abstract summary: We introduce COIN, a benchmark designed to assess interactive reasoning in realistic robotic manipulation.<n>First, we construct COIN-50: 50 interactive tasks in daily scenarios, and create COIN-Primitive required by causally-dependent tasks.<n>Second, we develop a low-cost mobile AR teleoperation system and collect the COIN-Primitive dataset with 50 demonstrations per primitive task.<n>Third, we develop systematic evaluation metrics about execution stability and generalization robustness to evaluate CodeAsPolicy, VLA, and language-conditioned H-VLA approaches.
• Score: 12.219913017710125
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Generalist embodied agents must perform interactive, causally-dependent reasoning, continually interacting with the environment, acquiring information, and updating plans to solve long-horizon tasks before they could be adopted in real-life scenarios. For instance, retrieving an apple from a cabinet may require opening multiple doors and drawers before the apple becomes visible and reachable, demanding sequential interaction under partial observability. However, existing benchmarks fail to systematically evaluate this essential capability. We introduce COIN, a benchmark designed to assess interactive reasoning in realistic robotic manipulation through three key contributions. First, we construct COIN-50: 50 interactive tasks in daily scenarios, and create COIN-Primitive required by causally-dependent tasks, and COIN-Composition with mid-term complexity for skill learning and generalization evaluation. Second, we develop a low-cost mobile AR teleoperation system and collect the COIN-Primitive Dataset with 50 demonstrations per primitive task (1,000 in total). Third, we develop systematic evaluation metrics about execution stability and generalization robustness to evaluate CodeAsPolicy, VLA, and language-conditioned H-VLA approaches. Our comprehensive evaluation reveals critical limitations in current methods: models struggle with interactive reasoning tasks due to significant gaps between visual understanding and motor execution. We provide fine-grained analysis of these limitations.

Related papers

• Inter-X++: A Comprehensive Benchmark for Multimodal Human-Human Interaction Analysis [55.78822230833247]
  We present Inter-X++, a comprehensive and large-scale benchmark designed to empower versatile HHI analysis.<n>Inter-X++ provides 11,388 high-fidelity interaction sequences and over 8.1M frames, featuring precise whole-body movements and detailed finger articulations.<n>We propose OpenHHI, a single and unified HHI representation and modeling framework that jointly optimize interaction reconstruction and semantic understanding.
  arXiv  Detail & Related papers  (2026-08-20T17:51:48Z)
• HRIBench: Benchmarking Interaction-Centric Human-Robot Collaboration [62.19372189311434]
  We introduce HRIBench, a diagnostic benchmark for intent-aware human-robot collaboration based on executable scenarios.<n>HRIBench represents collaborative tasks as structured scenario scripts that explicitly model agent roles, temporal dependencies, coordination constraints, and human behavior distributions.<n>Results show that current foundation robot policies struggle substantially in collaborative settings despite strong manipulation ability.
  arXiv  Detail & Related papers  (2026-07-05T09:15:35Z)
• EgoBench: An Interactive Egocentric Multimodal Benchmark for Tool-Using Agents [17.727481701114556]
  We introduce EgoBench, the first interactive multimodal benchmark for tool-using agents.<n>We implement a three-stage synergistic pipeline through which each task is designed to enforce the joint application of visual perception and tool-augmented multi-hop reasoning.<n>We additionally develop a multi-agent simulated user within EgoBench to evaluate agents' interaction capabilities, which generates high-fidelity, task-aligned responses.
  arXiv  Detail & Related papers  (2026-05-27T01:28:15Z)
• SocialOmni: Benchmarking Audio-Visual Social Interactivity in Omni Models [86.19617358080016]
  Social Omni is a benchmark that operationalizes the evaluation of conversational interactivity across three core dimensions.<n>Social Omni features 2,000 perception samples and a quality-controlled diagnostic set of 209 interaction-generation instances.<n>Our analysis reveals a pronounced decoupling between a model's perceptual accuracy and its ability to generate contextually appropriate interruptions.
  arXiv  Detail & Related papers  (2026-03-17T17:58:44Z)
• From Perception to Action: An Interactive Benchmark for Vision Reasoning [51.11355591375073]
  Causal Hierarchy of Actions and Interactions (CHAIN) benchmark designed to evaluate whether models can understand, plan, and execute structured action sequences grounded in physical constraints.<n> CHAIN shifts evaluation from passive perception to active problem solving, spanning tasks such as interlocking mechanical puzzles and 3D stacking and packing.<n>Our results show that top-performing models still struggle to internalize physical structure and causal constraints, often failing to produce reliable long-horizon plans and cannot robustly translate perceived structure into effective actions.
  arXiv  Detail & Related papers  (2026-02-24T15:33:02Z)
• IDRBench: Interactive Deep Research Benchmark [22.089706516440902]
  We introduce IDRBench, the first benchmark for systematically evaluating interactive deep research.<n>IDRBench combines a modular multi-agent research framework with on-demand interaction, a scalable reference-grounded user simulator, and an interaction-aware evaluation suite.
  arXiv  Detail & Related papers  (2026-01-10T20:29:12Z)
• Beyond Description: Cognitively Benchmarking Fine-Grained Action for Embodied Agents [52.14392337070763]
  We introduce CFG-Bench, a new benchmark designed to systematically evaluate fine-grained action intelligence.<n>CFG-Bench consists of 1,368 curated videos paired with 19,562 three-modalities question-answer pairs targeting four cognitive abilities.<n>Our comprehensive evaluation on CFG-Bench reveals that leading MLLMs struggle to produce detailed instructions for physical interactions.
  arXiv  Detail & Related papers  (2025-11-24T02:02:29Z)
• Interactive Evaluation of Large Language Models for Multi-Requirement Software Engineering Tasks [15.072898489107887]
  We build on DevAI, a benchmark of 55 programming tasks, by adding ground-truth solutions and evaluating the relevance and utility of interviewer hints.<n>Our results highlight the importance of dynamic evaluation in advancing the development of collaborative code-generating agents.
  arXiv  Detail & Related papers  (2025-08-26T10:22:37Z)
• EOC-Bench: Can MLLMs Identify, Recall, and Forecast Objects in an Egocentric World? [52.99661576320663]
  multimodal large language models (MLLMs) have driven breakthroughs in egocentric vision applications.<n>EOC-Bench is an innovative benchmark designed to systematically evaluate object-centric embodied cognition in dynamic egocentric scenarios.<n>We conduct comprehensive evaluations of various proprietary, open-source, and object-level MLLMs based on EOC-Bench.
  arXiv  Detail & Related papers  (2025-06-05T17:44:12Z)
• EgoNormia: Benchmarking Physical Social Norm Understanding [52.87904722234434]
  EGONORMIA spans seven norm categories: safety, privacy, proxemics, politeness, cooperation, coordination/proactivity, and communication/legibility.<n>Our work demonstrates that current state-of-the-art vision-language models (VLMs) lack robust grounded norm understanding, scoring a maximum of 54% on EGONORMIA and 65% on EGONORMIA-verified.
  arXiv  Detail & Related papers  (2025-02-27T19:54:16Z)
• TestAgent: Automatic Benchmarking and Exploratory Interaction for Evaluating LLMs in Vertical Domains [19.492393243160244]
  Large Language Models (LLMs) are increasingly deployed in highly specialized vertical domains.<n>Existing evaluations for vertical domains typically rely on the labor-intensive construction of static single-turn datasets.<n>We propose TestAgent, a framework for automatic benchmarking and exploratory dynamic evaluation in vertical domains.
  arXiv  Detail & Related papers  (2024-10-15T11:20:42Z)
• What to Say and When to Say it: Live Fitness Coaching as a Testbed for Situated Interaction [5.958765450103163]
  We present the QEVD benchmark and dataset, which explores human-AI interaction in the challenging, yet controlled, real-world domain of fitness coaching.<n>The benchmark requires vision-language models to recognize complex human actions, identify possible mistakes, and provide appropriate feedback in real-time.<n>Motivated by this, we propose a simple end-to-end streaming baseline that can respond asynchronously to human actions with appropriate feedback at the appropriate time.
  arXiv  Detail & Related papers  (2024-07-11T00:10:45Z)
• Cascaded Human-Object Interaction Recognition [175.60439054047043]
  We introduce a cascade architecture for a multi-stage, coarse-to-fine HOI understanding. At each stage, an instance localization network progressively refines HOI proposals and feeds them into an interaction recognition network. With our carefully-designed human-centric relation features, these two modules work collaboratively towards effective interaction understanding.
  arXiv  Detail & Related papers  (2020-03-09T17:05:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.