CORTEXA
← Browse
arxivcs.AI2026-07-10

ARCANA: A Reflective Multi-Agent Program Synthesis Framework for ARC-AGI-2 Reasoning

Kunbo Zhang, Lei Fu, Zeyu Wang, Zijing Liu, Kejian Tong

We present ARCANA, a collaborative multi agent framework for solving ARC AGI 2 tasks under strict test time and hardware constraints. ARCANA decomposes each task into iterative perception, hypothesis generation, symbolic execution, and reflective refinement. A perceptual grounding agent builds object centric scene graphs from raw grids, a latent program policy proposes diverse DSL programs, a symbolic executor verifies candidates on demonstrations, and a reflective agent synthesizes failure driven feedback for the next turn. These agents communicate through a shared differentiable blackboard and are scheduled by a learned meta controller. The design combines structured program search with adaptive multi turn correction, improving reasoning efficiency and solution quality on challenging abstract transformation tasks.

View free PDFSource page

Related papers

arxivcs.AI2026-07-16

Precise but Uncoupled: Reviewer Precision Does Not Guarantee Critique Uptake in Multi-Agent Math Reasoning

Chih-Hsuan Yang, Jingyan Jiang, Vikram Vasudevan, Cheng-Hau Yang, Huihuo Zheng, Le Chen, et al.

Many math- and science-oriented agent systems use hierarchical designs with specialized reviewer roles, assuming that a dedicated review stage should help turn wrong candidates into correct ones. We test this assumption on 4,181 verifier-grounded Omni-MATH problems using matched…

View free PDFSource page
arxivcs.AI2026-07-10

L-MAD: A Systematic Evaluation of Multi-Agent Debate Structures in Legal Reasoning

Tan-Minh Nguyen, Hoang-Trung Nguyen, Huu-Dong Nguyen, Dinh-Truong Do, Thi-Hai-Yen Vuong, Le-Minh Nguyen

While multi-agent debate (MAD) frameworks have shown significant potential in general reasoning, their effectiveness in highly structured, knowledge-heavy legal domains remains under-explored. In this work, we introduce the Legal Multi-Agent Debate (L-MAD) framework to systematic…

View free PDFSource page
arxivcs.MAcs.AI2026-07-21

Strategy-Following Multi-Agent Deep Reinforcement Learning Considering Control Strategies Provided to Other Agents

Yamato Takahagi, Gentoku Nakasone, Yoshinari Motokawa, Toshiharu Sugawara

This study proposes a learning method for multi-agent systems that allows agents to be controlled through human manager instructions after learning and enables uninstructed agents to implicitly complement the overall work based on the actions of other agents. Multi-agent applicat…

View free PDFSource page
arxivcs.AIcs.ETcs.MA2026-07-17

AgentFAIR: A Multi-Agent Collaborative Framework for FAIRness Evaluation of Geospatial Datasets

Ming Chen, Pranav Pai

Geospatial datasets support applications from urban planning to climate modeling, yet consistent assessment of FAIR compliance is difficult. Existing evaluators use different rubrics and evidence sources and may fail on JavaScript-rendered pages or repository-specific identifiers…

View free PDFSource page
arxivcs.CRcs.AIcs.MA2026-07-20

ChannelGuard: Safe Models Do Not Compose into Safe Multi-Agent Systems

Elias Hossain, Md Mehedi Hasan Nipu, Fatema Tuj Johora Faria, Tasfia Nuzhat Ornee, Maleeha Sheikh

Multi-agent LLM applications chain a planner, worker agents, a verifier, and a synthesizer, and every hop between agents is an unmonitored channel through which an adversary can smuggle instructions. Existing defenses guard only the input boundary (IBProtector, Llama Guard, perpl…

View free PDFSource page