Research
Papers, notes, talks, and reproducible work.
Based on long-task use, this paper examines how an Agent can adapt state, evidence, and action feedback to uncertainty, waiting, failure, and takeover needs, while documenting how delegated execution changes human pressure, trust, and capability boundaries.
Read summaryBased on ChatLab design work, this paper records how presence, response, waiting, and delivery patterns from human collaboration can make Agent runtime states easier to understand; it is an HCI design observation about feedback, evidence, intervention, and result confirmation.
Read summaryThis paper examines task decomposition, subtask capability classification, model selection, and bounded execution-time fallback, while reducing the broader Agent scheduling problem to an engineering slice of dsh-quota-router.
Read summaryA framework paper that advances structured thinking from the generic expression of "conclusion first, mutually exclusive and collectively exhaustive" into a four-stage problem-location method — Structure → Surface connections → Infer root problem → Find solution — with the causal, duality, dialectical, position, and interest lenses at its core, offering a testable training protocol and its boundaries of use.
Read summaryBased on the official second-hand residential price indices of 70 cities (2019–2026, 90 consecutive months with no gaps), we propose an interpretable three-component combination forecast (trend continuation / mean reversion / seasonality, with weights selected by rolling out-of-sample validation), provide five-year forecasts for the nation and Beijing, and discuss the possibility of a bottom.
Read summaryA theoretical synthesis, conceptual model, and application framework: it redefines motivation from "whether someone has drive" into a situated process of "whether motivation is generated, whether it can be converted into behavior, and whether it can be sustained," integrating self-determination theory, expectancy-value, goal setting, self-efficacy, and behavior models; bringing in perspectives from developmental psychology, business management, and organizational and occupational psychology to propose a three-layer "Situational Motivation Model" with a "growth axis," which it concretizes in the workplace into six components — task, goal, incentive, feedback, development path, and leadership/relationship — connecting to measurable outcomes such as engagement, commitment, performance, burnout, and turnover, introducing the enterprise life cycle as an enterprise-level moderating variable, and ultimately yielding nine falsifiable laws of motivation.
Read summaryA position paper based on the author's experience driving AI infrastructure and pilots in a real engineering organization: it proposes four kinds of boundaries—context, responsibility, authorization, and measurement—and offers a falsifiable pilot protocol plus a minimal harness as a demonstration of putting them into practice.
Read summaryA research-design and protocol paper: reframing data measurement from what belongs on the dashboard into a recomputable, reviewable, decision-supporting collaboration protocol, with minimal mechanisms for definitions, measurement units, metric tiering, dictionaries, and retrospectives.
Read summaryA research synthesis and position paper: defining engineering productivity as a continuously shortening "propose change — get trustworthy feedback — correct safely" loop, with metrics, a default path, and a falsifiable pilot protocol.
Read summaryA position paper: functional lines were efficient in an era of highly specialized work; once tools lower skill barriers, organizations should be redesigned around end-to-end business capability. It provides the mechanism, a spectrum of organizational forms, judgment signals, the two Chinese and American starting points, talent needs, and a falsifiable pilot protocol.
Read summaryA research-design and protocol paper: it proposes a material-organization and risk-annotation protocol aimed at safety actions, covering intimate and general interpersonal relationships, and defines AI's role as a structured co-pilot.
Read summaryA research-design and protocol paper proposing a safety-action-oriented framework for organizing material and annotating risk in workplace settings, covering top-middle-bottom three-tier safety, power structures, horizontal competition, authority-responsibility misalignment, career safety, career development, and occupational mental health.
Read summaryHow to build reproducible China housing price data across official 70-city indexes, Beijing district-level listings, LPR, and transaction records without confusing incompatible measures.
Read summaryA system design paper: modeling long-form fiction collaboration as recoverable, auditable, author-approved state transitions, and proposing an evaluation scheme that does not mistake mechanical scores for literary value.
Read summary