arXiv subjects

Jinling Gan

Publications and source records attributed to Jinling Gan.

3 recordsLinked to original sources

arXiv2025

Prepared mind, fast response: A temporal decoupling framework for adaptive knowledge orchestration in open-domain dialogue

The latency-quality tradeoff is a fundamental constraint in open-domain dialogue AI systems, since comprehensive knowledge access necessitates prohibitive response delays. Contemporary approaches offer two inadequate solutions: lightweight instruct models achieve sub-second latency but lack reasoning depth, while tool-augmented ReAct agents enhance factuality through external knowledge at the cost of synchronous execution that blocks interaction during retrieval processes. PMFR is thus proposed, with a temporal decoupling framework that fundamentally resolves the contradiction through asynchronous knowledge orchestration. PMFR employs three coordinated components: (1) a Knowledge Adequacy Evaluator for real-time sufficiency assessment, (2) a Lightweight Response Generator for immediate user interaction, and (3) an Asynchronous Knowledge Refinement Agent for background knowledge enhancement. This architecture maintains continuous conversational flow while progressively enriching knowledge coverage through intelligent triggering mechanisms. Evaluation results on TopiOCQA demonstrate PMFR outperforms brute-force scaling: PMFR achieves 95.3% latency reduction (23.38s -> 1.09s) while preserving response quality comparable to heavyweight synchronous baselines (GEval-C: 0.613 vs. 0.620).

cs.AI↗

arXiv2025

Multi-Agent Analysis of Off-Exchange Public Information for Cryptocurrency Market Trend Prediction

Cryptocurrency markets present unique prediction challenges due to their extreme volatility, 24/7 operation, and hypersensitivity to news events, with existing approaches suffering from key information extraction and poor sideways market detection critical for risk management. We introduce a theoretically-grounded multi-agent cryptocurrency trend prediction framework that advances the state-of-the-art through three key innovations: (1) an information-preserving news analysis system with formal theoretical guarantees that systematically quantifies market impact, regulatory implications, volume dynamics, risk assessment, technical correlation, and temporal effects using large language models; (2) an adaptive volatility-conditional fusion mechanism with proven optimal properties that dynamically combines news sentiment and technical indicators based on market regime detection; (3) a distributed multi-agent coordination architecture with low communication complexity enabling real-time processing of heterogeneous data streams. Comprehensive experimental evaluation on Bitcoin across three prediction horizons demonstrates statistically significant improvements over state-of-the-art natural language processing baseline, establishing a new paradigm for financial machine learning with broad implications for quantitative trading and risk management systems.

q-fin.CP↗

arXiv2025

COCO: Cognitive Operating System with Continuous Oversight for Multi-Agent Workflow Reliability

A critical limitation in large-scale multi-agent systems is the cascading of errors. And without intermediate verification, downstream agents exacerbate upstream inaccuracies, resulting in significant quality degradation. To bridge this gap, we introduce \textbf{COCO} (\textbf{C}ognitive \textbf{O}perating System with \textbf{C}ontinuous \textbf{O}versight), a theoretically grounded framework for asynchronous self-monitoring and adaptive error correction in multi-agent systems. COCO reconciles the fundamental tension between quality assurance and computational efficiency via a novel decoupled architecture. This design isolates error detection from the critical execution path and incorporates an automated configuration engine to minimize deployment complexity. The framework relies on three algorithmic innovations to mitigate both systematic and stochastic errors: (1) a Contextual Rollback Mechanism that leverages execution history for informed state recovery rather than naive retries; (2) a Bidirectional Reflection Protocol to ensure convergence and prevent oscillatory control loops; and (3) a Heterogeneous Cross-Validation Mechanism that utilizes ensemble disagreement to identify bias and hallucinations. Extensive experiments on diverse benchmarks demonstrate that COCO delivers a 6.5\% average performance improvement. Notably, the framework achieves 95.1\% of large-model performance with a 30$\times$ parameter reduction, confirming the potential for efficient, high-reliability deployment, and establishing COCO as a practical, annotation-based solution for critical autonomous domains.

cs.MA↗