Context: Proactive Goal-Directed Intelligence via Composable Sandboxed Programs, Declarative Wiring, and Structured Interaction

• URL: http://arxiv.org/abs/2605.23928v1
• Date: Tue, 21 Apr 2026 14:39:40 GMT
• Title: Context: Proactive Goal-Directed Intelligence via Composable Sandboxed Programs, Declarative Wiring, and Structured Interaction
• Abstract summary: We present Context, the intelligence layer of the Magarshak Architecture.<n>It replaces reactive query-response chatbots with proactive goal-directed agents that advance shared tasks without waiting for user prompts.<n>We prove six formal results: the Context Stability Theorem, bounding per-turn LM cost as a function of semantic change rate; a Program Composition Correctness Theorem; a Declarative Wiring Soundness Theorem; the Proactive Dominance Theorem.
• Score: 0.0
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present Context, the intelligence layer of the Magarshak Architecture, which replaces reactive query-response chatbots with proactive goal-directed agents that advance shared tasks without waiting for user prompts. The architecture rests on three mutually reinforcing mechanisms. Write-time context assembly precomputes enriched typed attributes via Groker agents, assembling interaction context as a deterministic pure function of graph state; context blocks are byte-identical across turns between semantic changes, enabling near-100% KV-cache reuse. Composable sandboxed wisdom programs form a governed library of LM-generated imperative programs declaratively wired to goal types via typed stream relations, composed via phase ordering, and executed at interaction time without further LM calls. Proactive goal stream state machines drive conversations toward terminal states by inspecting graph state and emitting structured interaction content (option arrays, governance affordances, clarification prompts) without awaiting user input. We prove six formal results: the Context Stability Theorem, bounding per-turn LM cost as a function of semantic change rate; a Program Composition Correctness Theorem; a Declarative Wiring Soundness Theorem; the Proactive Dominance Theorem, proving proactive agents weakly dominate reactive agents on expected turns-to-terminal-state; Coordination Overhead Elimination and Quality Preservation, establishing Pareto improvements in multi-participant goal chats; and a Cross-Platform Vote Consistency Theorem. Implemented in the open-source Qbix / Safebox / Safebots stack.

Related papers

• Inference-Time Graph Engineering for Multi-Agent LLM Workflows [51.56038180639833]
  We introduce ReActNet, a training-free framework that compiles a query and a set of role-specialized agents into a sequence of directed communication graphs.<n>We show that ReActNet consistently improves over fixed-topology and learned-topology baselines while maintaining competitive inference cost.
  arXiv  Detail & Related papers  (2026-09-04T23:32:11Z)
• EMERGE-Policy: A Robot Mind Emerges Beyond a Single Policy [48.898967774856466]
  EMERGE-Policy coordinates both capability invocation and information exchange.<n>Main Agent retains task-level state within an active context window.<n>Role-specific Sub Agents process perception, execution monitoring, verification, and memory consolidation.
  arXiv  Detail & Related papers  (2026-08-30T16:46:06Z)
• ScrambleToolBench: Agents Search Exhaustively Even When Their Own Map Points to the Next Step [74.83535434786145]
  We introduce ScrambleToolBench, an interactive terminal benchmark designed to isolate behavioral reasoning.<n>The benchmark requires agents to uncover hidden tool behaviors entirely through trial-and-error interaction.<n>Our evaluation of state-of-the-art language models reveals that successful initial discovery does not translate into robust adaptation.
  arXiv  Detail & Related papers  (2026-08-03T15:07:46Z)
• From Chatbot to Digital Colleague: The Paradigm Shift Toward Persistent Autonomous AI [74.72113315839412]
  Large Language Models (LLMs) are undergoing a fundamental transformation into integrated AI systems capable of reasoning, memory, and self-improvement.<n>We conceptualize this transition as a shift from episodic to Digital Colleague: from conversational answers to persistent work.
  arXiv  Detail & Related papers  (2026-06-12T14:33:55Z)
• Grokers: Bottom-Up Inductive Comprehension and Write-Time Intelligence over Typed Knowledge Graphs [0.0]
  Grokers is an architecture for building persistent, structured comprehension of typed knowledge graphs.<n> autonomous Groker agents analyze nodes in a typed stream graph, extract structured attributes via governed language model calls.
  arXiv  Detail & Related papers  (2026-05-07T17:28:36Z)
• IMPACT-Scribe: Interactive Temporal Action Segmentation with Boundary Scribbles and Query Planning [75.72434806468884]
  IMPACT-Scribe is a correction-driven framework for dense labeling.<n>It combines uncertainty-aware boundary supervision, local proposal modeling, cost-aware query planning, structured propagation, and correction-driven adaptation.<n> Experiments and a human study show that this closed-loop design improves labeling quality per effort, enhances boundary accuracy, and fosters better humanmachine interaction over time.
  arXiv  Detail & Related papers  (2026-05-03T01:45:58Z)
• The Auton Agentic AI Framework [5.410458076724158]
  The field of Artificial Intelligence is undergoing a transition from Generative AI to Agentic AI.<n>This transition exposes a fundamental architectural mismatch: Large Language Models (LLMs) produce unstructured outputs, whereas the backend infrastructure they must control requires deterministic, schema-conformant inputs.<n>The present paper describes the Auton Agentic AI Framework, a principled architecture for the creation, creation, and governance of autonomous agent.
  arXiv  Detail & Related papers  (2026-02-27T06:42:08Z)
• Factored Reasoning with Inner Speech and Persistent Memory for Evidence-Grounded Human-Robot Interaction [38.120218754079055]
  We introduce JANUS, a cognitive architecture for assistive robots that models interaction as a partially observable Markov decision process.<n>Janus decomposes the overall behavior into specialized modules, related to scope detection, intent recognition, memory, inner speech, query generation, and outer speech.<n>We evaluate JANUS through module-level unit tests in a dietary assistance domain grounded on a knowledge graph, reporting high agreement with curated references and practical latency profiles.
  arXiv  Detail & Related papers  (2026-01-31T11:42:33Z)
• From Transcripts to AI Agents: Knowledge Extraction, RAG Integration, and Robust Evaluation of Conversational AI Assistants [0.0]
  Building reliable conversational AI assistants for customer-facing industries remains challenging due to noisy conversational data, fragmented knowledge, and the requirement for accurate human hand-off.<n>This paper presents an end-to-end framework for constructing and evaluating a conversational AI assistant directly from historical call transcripts.
  arXiv  Detail & Related papers  (2026-01-26T07:44:47Z)
• From Completion to Editing: Unlocking Context-Aware Code Infilling via Search-and-Replace Instruction Tuning [81.97788535387286]
  We propose a framework that internalizes the agentic verification-and-editing mechanism into a unified, single-pass inference process.<n>With minimal data, SRI-Coder enables Chat models to surpass the completion performance of their Base counterparts.<n>Unlike FIM-style tuning, SRI preserves general coding competencies and maintains inference latency comparable to standard FIM.
  arXiv  Detail & Related papers  (2026-01-19T20:33:53Z)
• Rethinking Testing for LLM Applications: Characteristics, Challenges, and a Lightweight Interaction Protocol [83.83217247686402]
  Large Language Models (LLMs) have evolved from simple text generators into complex software systems that integrate retrieval augmentation, tool invocation, and multi-turn interactions.<n>Their inherent non-determinism, dynamism, and context dependence pose fundamental challenges for quality assurance.<n>This paper decomposes LLM applications into a three-layer architecture: textbftextitSystem Shell Layer, textbftextitPrompt Orchestration Layer, and textbftextitLLM Inference Core.
  arXiv  Detail & Related papers  (2025-08-28T13:00:28Z)
• Scalable In-Context Q-Learning [68.9917436397079]
  We propose textbfScalable textbfIn-textbfContext textbfQ-textbfLearning (textbfSICQL) to steer in-context reinforcement learning.<n>textbfSICQL harnesses dynamic programming and world modeling to steer ICRL toward efficient reward and task generalization.
  arXiv  Detail & Related papers  (2025-06-02T04:21:56Z)
• Modular Speaker Architecture: A Framework for Sustaining Responsibility and Contextual Integrity in Multi-Agent AI Communication [0.0]
  Sustaining coherent, role-aware communication across multi-agent systems remains a foundational challenge in AI.<n>We propose the Modular Speaker Architecture (MSA), a framework that decomposes speaker behavior into modular components for role tracking, responsibility continuity, and contextual coherence.
  arXiv  Detail & Related papers  (2025-06-01T17:39:51Z)
• QLASS: Boosting Language Agent Inference via Q-Guided Stepwise Search [89.97082652805904]
  We propose QLASS (Q-guided Language Agent Stepwise Search), to automatically generate annotations by estimating Q-values.<n>With the stepwise guidance, we propose a Q-guided generation strategy to enable language agents to better adapt to long-term value.<n>We empirically demonstrate that QLASS can lead to more effective decision making through qualitative analysis.
  arXiv  Detail & Related papers  (2025-02-04T18:58:31Z)
• Compromising Embodied Agents with Contextual Backdoor Attacks [69.71630408822767]
  Large language models (LLMs) have transformed the development of embodied intelligence. This paper uncovers a significant backdoor security threat within this process. By poisoning just a few contextual demonstrations, attackers can covertly compromise the contextual environment of a black-box LLM.
  arXiv  Detail & Related papers  (2024-08-06T01:20:12Z)

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.