Back to Details

creative-thinking-for-research

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Creative Thinking for Research

科研创造性思维指南

Eight empirically grounded frameworks from cognitive science, applied to computer science and AI research. Unlike ad-hoc brainstorming, each framework here is backed by decades of creativity research — from Koestler's bisociation to Kauffman's adjacent possible. They target distinct cognitive operations: combining, reformulating, analogizing, constraining, inverting, abstracting, exploring boundaries, and holding contradictions.
本文整理了8个来自认知科学、经实证验证的框架，可应用于计算机科学与人工智能研究场景。与临时头脑风暴不同，这里的每个框架都有几十年的创造力研究作为支撑——从Koestler的双重联想（bisociation）到Kauffman的相邻可能（adjacent possible），分别对应不同的认知操作：组合、重构、类比、约束调整、反转、抽象、边界探索以及矛盾兼容。

When to Use This Skill

适用场景

  • Generating genuinely novel ideas, not incremental extensions of prior work
  • Feeling trapped in a local optimum of thinking within a single subfield
  • Wanting to systematically apply creativity heuristics rather than waiting for inspiration
  • Preparing for a research retreat or PhD-level ideation session
  • Bridging between fields and seeking structural (not superficial) connections
Do NOT use this skill when:
  • You need structured project-level brainstorming workflows (use 
    brainstorming-research-ideas
    )
  • You have a well-defined problem and need execution help (use domain-specific skills)
  • You need a literature survey (use 
    scientific-skills:literature-review
    )
Relationship to Brainstorm skill: The brainstorm skill provides operational workflows (diverge → converge → refine) and practical filters. This skill provides the deeper cognitive engines that power creative leaps. Use them together: creative-thinking to generate raw insight, brainstorm to structure and evaluate it.
  • 生成真正具备创新性的想法，而非对已有工作的增量延伸
  • 陷入单一细分领域的思维局部最优，无法跳脱局限
  • 希望系统性应用创意启发方法，而非被动等待灵感降临
  • 筹备研究 retreat 或博士级别的创意构思会议
  • 希望建立跨领域的结构性（而非表层）关联
请勿在以下场景使用：
  • 需要结构化的项目级头脑风暴工作流（请使用 
    brainstorming-research-ideas
    ）
  • 已有明确定义的问题，需要执行层面的帮助（请使用对应领域的专项工具）
  • 需要做文献调研（请使用 
    scientific-skills:literature-review
    ）
与头脑风暴工具的关系：头脑风暴工具提供操作工作流（发散→收敛→打磨）和实用筛选规则，本工具提供支撑创意跃迁的深层认知引擎。两者可以搭配使用：用创造性思维生成原始洞见，用头脑风暴做结构化梳理和评估。

Framework 1: Combinatorial Creativity (Bisociation)

框架1：组合式创造力（双重联想 Bisociation）

Novel ideas arise from combining existing concepts in unexpected ways. Arthur Koestler called this bisociation — connecting two previously unrelated frames of reference, as distinct from routine association within a single frame.
Why it works: Meta-research consistently shows that breadth of knowledge is a precursor to creative output. People who read across disciplines produce more novel work. The combination itself is the creative act.
In CS Research:
  • Biological evolution → optimization (genetic algorithms)
  • Game theory → networking (mechanism design for routing)
  • Statistical physics → machine learning (Boltzmann machines, energy-based models)
  • Linguistics → programming (type theory, formal grammars)
Systematic Bisociation Workflow:
  1. Select two domains you have at least passing familiarity with
  2. List core primitives in each domain (5-10 fundamental concepts per domain)
  3. Create a cross-product matrix: row = concepts from Domain A, column = concepts from Domain B
  4. For each cell, ask: "What would it mean to apply A's concept to B's problem?"
  5. Filter: Which combinations produce a non-trivial, testable research question?
  6. Validate structural depth: Is the connection mechanistic or merely metaphorical?
Cross-Product Example:
CachingLoad BalancingFault Tolerance
Natural SelectionEvict least-fit entriesAdaptive allocation via fitnessPopulation-level redundancy
Immune MemoryLearned threat signaturesDistributed detectionSelf/non-self discrimination
SymbiosisCooperative prefetchingMutualistic resource sharingCo-dependent resilience
Quality Test: A strong bisociation is not a surface metaphor ("the network is like a brain") but a structural mapping where the mechanism transfers ("attention mechanisms implement a form of selective gating analogous to cognitive attention filtering").
Self-Check:
  • Is the connection structural (mechanisms map) or merely verbal (labels map)?
  • Does the combination generate testable predictions?
  • Would an expert in both fields find the connection non-obvious but sound?
新颖的想法来自于将现有概念以出人意料的方式组合。Arthur Koestler 将这种模式称为双重联想（bisociation）：连接两个此前毫无关联的参考框架，和单一框架内的常规联想有本质区别。
底层逻辑：元研究反复证明，知识广度是创意产出的前置条件。跨领域阅读的研究者能产出更多创新性成果，组合行为本身就是创意行为。
在CS研究中的典型案例：
  • 生物进化 → 优化算法（遗传算法）
  • 博弈论 → 网络技术（路由机制设计）
  • 统计物理 → 机器学习（玻尔兹曼机、能量模型）
  • 语言学 → 编程技术（类型论、形式文法）
系统性双重联想工作流：
  1. 选择两个你至少有基础了解的领域
  2. 列出每个领域的核心原语（每个领域列5-10个基础概念）
  3. 构建交叉乘积矩阵：行=领域A的概念，列=领域B的概念
  4. 针对每个矩阵单元，提问：「把A的概念应用到B的问题上，会产生什么效果？」
  5. 筛选：哪些组合能生成有价值、可验证的研究问题？
  6. 验证结构性深度：关联是机制层面的，还是仅停留在隐喻层面？
交叉乘积示例：
缓存负载均衡容错
自然选择淘汰适配度最低的条目基于适配度的自适应分配种群级别的冗余
免疫记忆已学习的威胁特征库分布式检测自我/非自我识别
共生关系协同预取互利性资源共享共弹性
质量检验标准：优质的双重联想不是表层隐喻（「网络就像大脑」），而是可迁移机制的结构性映射（「注意力机制实现了一种选择性门控，和认知注意力过滤的机制类似」）。
自检清单：
  • 关联是结构性的（机制可映射）还是仅文字层面的（标签可映射）？
  • 组合能否生成可验证的预测？
  • 两个领域的专家是否会认为这种关联虽不直观但逻辑自洽？

Framework 2: Problem Reformulation (Representational Change)

框架2：问题重构（表征转换）

Gestalt psychologists identified that breakthroughs often come not from solving the problem as stated, but from re-representing the problem itself. Kaplan and Simon's work on insight shows that changing the problem space — the constraints, the abstraction level, the formalism — is often where creativity lives.
The Key Shift: From "How do I solve this problem?" to "Am I even thinking about this problem correctly?"
Reformulation Strategies:
StrategyExample
Change the objective"Make the algorithm faster" → "Eliminate the need for this computation"
Change the formalismGraph problem → linear algebra problem (spectral methods)
Change the granularityPer-token prediction → per-span prediction
Change the agent"How should the model learn?" → "How should the data teach?" (curriculum learning)
Change the timescaleReal-time optimization → amortized inference
Invert the directionForward simulation → inverse problem (learning from observations)
Workflow:
  1. State your current problem in one sentence
  2. Identify the hidden assumptions in that statement:
    • What formalism are you using? (Could you use a different one?)
    • What is the objective? (Is it the right objective?)
    • What level of granularity? (Could you go coarser or finer?)
    • Who is the agent? (Could you shift perspective?)
  3. For each assumption, generate the alternative: "What if [opposite assumption]?"
  4. For each alternative, ask: "Does this reformulation make the problem easier, harder, or different in a useful way?"
  5. A reformulation that makes a hard problem easy is often a publishable insight on its own
Classic CS Examples:
  • PageRank: Reformulated "find important web pages" from content analysis to graph eigenvalue problem
  • Dropout: Reformulated "prevent overfitting" from regularization to approximate ensemble
  • Attention: Reformulated "handle long sequences" from remembering everything to selectively querying
格式塔心理学家发现，突破往往不是来自解决提出的问题，而是来自对问题本身进行重新表征。Kaplan和Simon关于洞察力的研究表明，改变问题空间——约束条件、抽象层级、形式化表达——往往是创意的来源。
核心思维转变：从「我该怎么解决这个问题？」转向「我对这个问题的理解本身是不是正确的？」
重构策略：
策略示例
改变目标「让算法更快」→ 「彻底消除这项计算的需求」
改变形式化表达图问题 → 线性代数问题（谱方法）
改变粒度逐token预测 → 逐span预测
改变主体「模型该怎么学习？」→ 「数据该怎么教授知识？」（课程学习）
改变时间尺度实时优化 → 摊销推理
反转方向前向仿真 → 逆问题（从观测结果学习）
工作流：
  1. 用一句话描述你当前的问题
  2. 识别描述中的隐含假设：
    • 你在用什么形式化表达？有没有其他可选的表达？
    • 目标是什么？是不是正确的目标？
    • 当前的粒度是多少？能不能更粗或者更细？
    • 行动主体是谁？能不能切换视角？
  3. 针对每个假设，生成替代方案：「如果[相反的假设]成立会怎么样？」
  4. 针对每个替代方案，提问：「这种重构有没有让问题变得更简单、更难，或是出现了有价值的新变化？」
  5. 能把难题变简单的重构本身往往就是可发表的洞见。
经典CS案例：
  • PageRank：把「找到重要网页」的问题从内容分析重构为图特征值问题
  • Dropout：把「防止过拟合」的问题从正则化重构为近似集成
  • Attention：把「处理长序列」的问题从记忆全量信息重构为选择性查询

Framework 3: Analogical Reasoning (Structure-Mapping)

框架3：类比推理（结构映射）

Dedre Gentner's structure-mapping theory and Kevin Dunbar's studies of real scientists show that analogy is the core engine of scientific creativity. The critical finding: surface-level analogies are common but weak; structural or relational analogies — where the deep causal/relational structure maps across domains — produce the most powerful insights.
Dunbar's Finding: In the most successful labs, analogies from distant domains drove the most important discoveries. Nearby analogies refined ideas; distant analogies generated them.
Levels of Analogical Depth:
LevelDescriptionValueExample
SurfaceThings look similarLow"A neural network is like a brain"
RelationalRelationships between entities matchMedium"Attention allocation in models parallels resource allocation in economics"
StructuralDeep causal mechanisms mapHigh"Diffusion models reverse a thermodynamic process; the math of non-equilibrium stat-mech directly applies"
Structure-Mapping Workflow:
  1. Describe your problem using only relational/causal language (strip domain-specific nouns)
    • Bad: "We need to improve transformer attention efficiency"
    • Good: "We have a system that must selectively aggregate information from a large set, where relevance is context-dependent and the cost scales quadratically with set size"
  2. Search for structural matches: What other systems selectively aggregate from large sets?
    • Database query optimization, visual attention in neuroscience, information retrieval, resource allocation
  3. Pick the most distant match with genuine structural fidelity
  4. Map the solution mechanism: How does the source domain solve this?
  5. Transfer and adapt: What changes when you bring that mechanism into your domain?
  6. Generate predictions: The analogy should tell you something you didn't already know
Validation Checklist:
  • Does the mapping preserve causal/relational structure (not just labels)?
  • Can I identify at least one prediction the analogy makes in my domain?
  • Would an expert in the source domain confirm the mechanism is correctly understood?
  • Is the analogy non-obvious to my target audience?
Dedre Gentner的结构映射理论和Kevin Dunbar对真实科学家的研究表明，类比是科学创造力的核心引擎。关键结论是：表层类比很常见但价值很低；结构或关系类比——深层因果/关系结构可跨领域映射——才能产生最有价值的洞见。
Dunbar的研究发现：在最成功的实验室中，来自遥远领域的类比推动了最重要的发现。邻近类比用于打磨想法，遥远类比用于生成新想法。
类比深度层级：
层级描述价值示例
表层事物看起来相似低「神经网络就像大脑」
关系层实体之间的关系匹配中「模型中的注意力分配和经济学中的资源分配逻辑相似」
结构层深层因果机制可映射高「Diffusion模型逆转了热力学过程，非平衡统计力学的数学方法可以直接应用」
结构映射工作流：
  1. 仅用关系/因果语言描述你的问题（剥离领域特定名词）
    • 反面示例：「我们需要提升Transformer注意力的效率」
    • 正面示例：「我们的系统需要从大规模集合中选择性聚合信息，相关性依赖上下文，且成本和集合规模呈二次方关系」
  2. 搜索结构匹配项：还有哪些系统需要从大规模集合中做选择性聚合？
    • 数据库查询优化、神经科学中的视觉注意力、信息检索、资源分配
  3. 选择结构保真度最高且最遥远的匹配项
  4. 映射解决方案机制：源领域是怎么解决这个问题的？
  5. 迁移和适配：把机制迁移到你的领域时需要做哪些调整？
  6. 生成预测：类比应该能给你带来之前不知道的新结论
验证清单：
  • 映射是否保留了因果/关系结构（而不只是标签）？
  • 我能不能至少找到一个类比在本领域生成的预测？
  • 源领域的专家是否会确认我对机制的理解是正确的？
  • 这个类比对目标受众来说是不是不那么直观？

Framework 4: Constraint Manipulation (Boden's Framework)

框架4：约束操作（Boden框架）

Margaret Boden's framework distinguishes three forms of creativity based on how they interact with constraints:
TypeOperationCS Example
ExploratorySearch within the existing conceptual spaceHyperparameter tuning, architecture search within a fixed paradigm
CombinationalCombine elements from different spacesMulti-task learning, neuro-symbolic methods
TransformationalChange the rules of the space itselfDropping the assumption that training requires labels (self-supervised learning)
Transformational creativity is the rarest and highest-impact. It happens when you change what is even considered a valid solution.
Constraint Analysis Workflow:
  1. List the constraints of your current approach (5-10 constraints):
    • Computational: "Must fit in GPU memory"
    • Methodological: "Requires labeled data"
    • Architectural: "Uses fixed-length context"
    • Evaluative: "Measured by accuracy on benchmark X"
  2. Classify each constraint:
    • Hard: Physically or logically necessary (cannot violate)
    • Soft: Convention or historical accident (can question)
    • Hidden: Not stated but implicitly assumed (most fertile for innovation)
  3. For each soft/hidden constraint, ask:
    • What if we relaxed it? (streaming algorithms from relaxing "fits in memory")
    • What if we tightened it? (efficiency research from tightening compute budgets)
    • What if we replaced it with a different constraint entirely?
  4. The most productive move is often exposing and dropping a hidden constraint
Classic Examples of Constraint Transformation:
  • "Data must fit in memory" → dropped → streaming algorithms, external memory
  • "Training requires human labels" → dropped → self-supervised learning
  • "Models must be deterministic" → dropped → variational methods, diffusion
  • "Inference must happen in one pass" → dropped → iterative refinement, chain-of-thought
Margaret Boden的框架根据和约束的交互方式，将创造力分为三类：
类型操作CS示例
探索型在现有概念空间内搜索超参数调优、固定范式内的架构搜索
组合型组合不同空间的元素多任务学习、神经符号方法
变革型改变空间本身的规则放弃训练需要标签的假设（自监督学习）
变革型创造力是最稀有、影响力最大的类型，它会改变「什么才是有效解决方案」的认知边界。
约束分析工作流：
  1. 列出你当前方法的约束条件（5-10个）：
    • 计算层面：「必须适配GPU显存」
    • 方法层面：「需要标注数据」
    • 架构层面：「使用固定长度上下文」
    • 评估层面：「用基准测试X的准确率衡量」
  2. 给每个约束分类：
    • 硬约束：物理或逻辑上必须满足（无法违反）
    • 软约束：惯例或历史遗留（可以质疑）
    • 隐约束：没有明确说明但默认假设的（最容易产生创新）
  3. 针对每个软约束/隐约束，提问：
    • 如果放松约束会怎么样？（放松「适配内存」的约束产生了流处理算法、外存算法）
    • 如果收紧约束会怎么样？（收紧计算预算产生了效率优化方向的研究）
    • 如果用完全不同的约束替换会怎么样？
  4. 最有成效的操作往往是暴露并放弃一个隐约束。
约束变革的经典案例：
  • 「数据必须适配内存」→ 放弃 → 流处理算法、外存算法
  • 「训练需要人工标注」→ 放弃 → 自监督学习
  • 「模型必须是确定性的」→ 放弃 → 变分方法、diffusion模型
  • 「推理必须单次完成」→ 放弃 → 迭代优化、思维链

Framework 5: Negation and Inversion

框架5：否定与反转

Take a core assumption in your field and negate it. This is formalized in De Bono's lateral thinking and the TRIZ methodology from engineering.
The Pattern: "What if [widely held assumption] is wrong, unnecessary, or invertible?"
Systematic Negation Workflow:
  1. List 5-10 core assumptions in your subfield (the things "everyone knows")
  2. Negate each one and ask: What system would you build?
  3. Evaluate each negation:
    • Incoherent → discard
    • Already explored → check if conditions have changed (see brainstorm skill, Framework 5)
    • Unexplored and coherent → potential research direction
Negation Hall of Fame in CS:
AssumptionNegationResult
"We need strong consistency"What if we don't?Eventual consistency, CRDTs
"We need exact answers"What if approximate is fine?Sketches, LSH, approximate nearest neighbors
"Labels are necessary"What if we learn without them?Self-supervised learning, contrastive methods
"More parameters = more compute"What if we don't use all parameters?Mixture of Experts, sparse models
"Training and inference are separate"What if the model keeps learning?Online learning, test-time training
"Errors must be prevented"What if we embrace and correct them?Speculative decoding, self-correction
TRIZ-Inspired Principles for CS:
TRIZ PrincipleCS Application
InversionReverse the process (generative vs. discriminative)
SegmentationBreak monolithic into modular (microservices, mixture of experts)
MergingCombine separate steps (end-to-end learning)
UniversalityOne component serves multiple functions (multi-task models)
NestingPlace one system inside another (meta-learning)
DynamizationMake static things adaptive (dynamic architectures, adaptive computation)
取你所在领域的一个核心假设，否定它。这是De Bono水平思考和工程领域TRIZ方法中的正式策略。
核心模式：「如果[广泛接受的假设]是错误的、不必要的，或是可以反转的，会怎么样？」
系统性否定工作流：
  1. 列出你所在细分领域的5-10个核心假设（所有人都默认成立的内容）
  2. 逐个否定，并提问：你能构建什么样的系统？
  3. 评估每个否定结果：
    • 逻辑不通 → 丢弃
    • 已经被研究过 → 检查前提条件是否发生了变化（参考头脑风暴工具的框架5）
    • 未被探索且逻辑自洽 → 潜在研究方向
CS领域的否定经典案例：
假设否定成果
「我们需要强一致性」如果不需要会怎么样？最终一致性、CRDT
「我们需要精确结果」如果近似结果足够用会怎么样？Sketches、LSH、近似最近邻
「标注是必要的」如果无标注学习会怎么样？自监督学习、对比学习方法
「参数越多计算量越大」如果不使用全部参数会怎么样？Mixture of Experts、稀疏模型
「训练和推理是分离的」如果模型持续学习会怎么样？在线学习、测试时训练
「必须避免错误」如果接纳并修正错误会怎么样？投机解码、自修正
适用于CS的TRIZ启发原则：
TRIZ原则CS应用
反转逆转流程（生成式 vs 判别式）
拆分把单体拆分为模块（微服务、Mixture of Experts）
合并合并独立步骤（端到端学习）
通用性一个组件支持多种功能（多任务模型）
嵌套把一个系统放在另一个系统内部（元学习）
动态化让静态内容自适应（动态架构、自适应计算）

Framework 6: Abstraction and Generalization Laddering

框架6：抽象与泛化阶梯

Moving up and down the abstraction ladder is a fundamental creative act. Polya's heuristics formalize this: "Can you solve a more general problem? A more specific one? An analogous one?"
Three Moves:
MoveQuestionOutcome
Generalize"Is my solution a special case of something broader?"Framework papers, unifying theories
Specialize"What happens when I add extreme constraints?"Niche applications, surprising edge cases
Analogize"Where else does this abstract pattern appear?"Cross-domain transfer (see Framework 3)
Generalization Workflow:
  1. State your specific result
  2. Replace each specific element with a variable: "ResNet works for ImageNet" → "Architecture X works for distribution Y"
  3. Ask: Under what conditions does this hold? What is the general principle?
  4. If the general principle is novel → that is the contribution
Specialization Workflow:
  1. Take a general method
  2. Add extreme constraints: tiny data, huge dimensionality, adversarial inputs, real-time requirements
  3. Ask: Does the method still work? If not, why not?
  4. The failure case often reveals the method's true assumptions
When to Generalize vs. Specialize:
  • Generalize when you have results but no explanation
  • Specialize when you have theory but no grounding
  • Analogize when you are stuck in either direction
在抽象阶梯上下移动是基础的创意行为。Polya的启发法将其正式化：「你能不能解决一个更通用的问题？或是更具体的问题？或是一个类似的问题？」
三种移动方式：
操作核心问题产出
泛化「我的解决方案是不是某个更通用体系的特例？」框架类论文、统一理论
特化「如果添加极端约束会发生什么？」细分场景应用、意外的边界 case
类比「这个抽象模式还会出现在哪些地方？」跨领域迁移（参考框架3）
泛化工作流：
  1. 陈述你的具体成果
  2. 把每个具体元素替换为变量：「ResNet在ImageNet上效果好」→ 「架构X在分布Y上效果好」
  3. 提问：这个结论在什么条件下成立？通用原理是什么？
  4. 如果通用原理是新颖的 → 这就是核心贡献
特化工作流：
  1. 取一个通用方法
  2. 添加极端约束：极小数据量、极高维度、对抗输入、实时要求
  3. 提问：方法还能正常工作吗？如果不能，原因是什么？
  4. 失败案例往往能暴露方法的真实隐含假设
泛化 vs 特化的选择：
  • 有结果但没有解释时选择泛化
  • 有理论但没有落地验证时选择特化
  • 两个方向都卡住时选择类比

Framework 7: The Adjacent Possible (Kauffman / Johnson)

框架7：相邻可能（Kauffman / Johnson）

Stuart Kauffman's concept, popularized by Steven Johnson: innovation happens at the boundary of what is currently reachable — the adjacent possible. New ideas become thinkable once their prerequisites exist. This explains why simultaneous independent discovery is so common — multiple people reach the same boundary.
Practical Implication: Map what has recently become possible and explore the space those enablers open.
Adjacent Possible Mapping Workflow:
  1. List recent enablers (last 1-3 years):
    • New hardware capabilities (longer context, faster inference, new accelerators)
    • New datasets or benchmarks
    • New open-source tools or frameworks
    • New theoretical results
    • New regulatory or social conditions
  2. For each enabler, ask: "What was previously impossible or impractical that this now permits?"
  3. Combine enablers: The most powerful adjacent possibles arise from the intersection of multiple new enablers
  4. Check for competition: If many people can see the same adjacent possible, speed or a unique angle matters
Current Adjacent Possibles (2025-2026):
EnablerNewly Possible
1M+ token context windowsFull-codebase reasoning, book-length analysis
Inference cost drops (100x in 2 years)Real-time agentic loops, always-on AI assistants
Open-weight models at GPT-4 levelReproducible research on frontier capabilities
Multimodal models (vision + language + audio)Unified perception-reasoning systems
Synthetic data at scaleTraining data for domains with no natural data
Tool-using modelsResearch automation, self-improving systems
Timing Signal: If your idea requires technology that doesn't exist yet, it's beyond the adjacent possible — park it. If your idea could have been done 5 years ago, someone probably did — check the literature. The sweet spot is ideas that became feasible in the last 6-18 months.
Stuart Kauffman提出、Steven Johnson推广的概念：创新发生在当前能力可触及的边界——相邻可能。当前置条件存在时，新想法才会出现。这也解释了为什么独立同时发现非常普遍：多个研究者同时抵达了同一个边界。
实践意义：梳理最近成为可能的技术，探索这些赋能工具打开的新空间。
相邻可能映射工作流：
  1. 列出近1-3年的新赋能工具：
    • 新硬件能力（更长上下文、更快推理、新加速器）
    • 新数据集或基准测试
    • 新开源工具或框架
    • 新理论成果
    • 新监管或社会条件
  2. 针对每个赋能工具，提问：「有什么之前不可能或不实际的事现在可以做了？」
  3. 组合多个赋能工具：最有价值的相邻可能往往来自多个新赋能工具的交叉
  4. 竞争情况检查：如果很多人都能看到同一个相邻可能，速度或独特视角就很重要
当前的相邻可能（2025-2026）：
赋能工具新可实现能力
1M+ token上下文窗口全代码库推理、整本书长度的内容分析
推理成本下降（2年降100倍）实时Agent循环、始终在线的AI助手
GPT-4级别的开源模型前沿能力的可复现研究
多模态模型（视觉+语言+音频）统一感知-推理系统
大规模合成数据无天然数据领域的训练数据
可使用工具的模型研究自动化、自改进系统
时机信号：如果你的想法需要还不存在的技术，说明超出了相邻可能，可以先搁置。如果你的想法5年前就可以实现，大概率已经有人做过了，先查文献。最佳窗口期是近6-18个月才变得可行的想法。

Framework 8: Janusian and Dialectical Thinking

框架8：两面神与辩证思维

Albert Rothenberg's studies of eminent creators found that holding two contradictory ideas simultaneously is a hallmark of creative thinking. Named after Janus, the two-faced Roman god, this mode of thinking doesn't resolve contradictions by choosing a side — it generates new frameworks that transcend the opposition.
In CS: The most influential results often emerge from tensions previously thought irreconcilable.
ContradictionResolutionImpact
Consistency AND Availability (distributed systems)CAP theorem: formalized the trade-off, then Raft/CRDTs found practical middle groundsFoundation of distributed systems theory
Security AND UsabilityZero-knowledge proofs: prove knowledge without revealing itEnabled private computation
Expressiveness AND TractabilityProbabilistic programming: express complex models, automate inferenceNew programming paradigm
Memorization AND GeneralizationGrokking: models memorize first, then generalize with more trainingNew understanding of learning dynamics
Compression AND QualityNeural codecs that compress beyond information-theoretic limits via learned priorsRedefined compression research
Dialectical Thinking Workflow:
  1. Identify a binary in your field: A vs. B (two approaches, goals, or paradigms treated as opposites)
  2. Resist choosing a side. Instead ask:
    • "What would a system look like that achieves both A and B?"
    • "Under what conditions is the A-B trade-off not fundamental?"
    • "Is the opposition an artifact of how we formalized the problem?"
  3. Seek synthesis: The resolution often requires a new abstraction that reframes the relationship
  4. Test the synthesis: Can you demonstrate empirically that both goals are achievable?
Self-Check:
  • Am I holding the contradiction genuinely (not prematurely resolving it)?
  • Is the synthesis a new idea, not just a compromise (splitting the difference)?
  • Does the resolution change how people think about the problem, not just the solution?
Albert Rothenberg对杰出创作者的研究发现，同时持有两个矛盾的想法是创造性思维的标志性特征。这个模式以罗马双面神Janus命名，它不会通过选边来解决矛盾——而是生成超越对立的新框架。
在CS领域：最有影响力的成果往往来自此前被认为不可调和的矛盾。
矛盾解决方案影响力
一致性 AND 可用性（分布式系统）CAP定理：形式化了权衡，之后Raft/CRDT找到了实用的中间方案分布式系统理论的基础
安全性 AND 易用性零知识证明：不泄露信息的前提下证明知识持有支撑了隐私计算
表达性 AND 可处理性概率编程：可表达复杂模型，自动化推理新编程范式
记忆 AND 泛化Grokking：模型先记忆，经过更多训练后实现泛化对学习动力学的新理解
压缩率 AND 质量神经编解码器通过学习先验突破信息论压缩极限重新定义了压缩研究
辩证思维工作流：
  1. 识别领域中的二元对立：A vs B（被视为对立的两种方法、目标或范式）
  2. 拒绝选边，转而提问：
    • 「有没有系统能同时实现A和B？」
    • 「在什么条件下A-B的权衡不是根本性的？」
    • 「这种对立是不是我们对问题形式化过程中的人为产物？」
  3. 寻求综合：解决方案往往需要一个新的抽象来重构两者的关系
  4. 验证综合方案：你能不能通过实证证明两个目标都可以实现？
自检清单：
  • 我是不是真的在持有矛盾（没有过早解决）？
  • 综合方案是不是新想法，而不只是妥协（各让一步）？
  • 解决方案是不是改变了人们对问题的认知，而不只是解决了问题？

Combining Frameworks: A Creative Thinking Protocol

框架组合：创造性思维执行协议

These frameworks are most powerful in combination. Here is a systematic protocol for a deep creative thinking session:
这些框架组合使用时效果最好。以下是深度创造性思维会议的系统性执行协议：

Phase 1: Map the Space (15 min)

阶段1：空间映射（15分钟）

  1. Constraint Manipulation (F4): List all constraints of the current paradigm. Mark which are hard, soft, hidden.
  2. Adjacent Possible (F7): List recent enablers that change the feasibility landscape.
  1. 约束操作（框架4）：列出当前范式的所有约束，标记硬约束、软约束、隐约束
  2. 相邻可能（框架7）：列出改变可行性边界的近期赋能工具

Phase 2: Generate Disruptions (30 min)

阶段2：生成突破点（30分钟）

  1. Negation (F5): Negate 3 soft/hidden constraints. What systems emerge?
  2. Bisociation (F1): Pick a distant field and create a cross-product matrix with your domain.
  3. Problem Reformulation (F2): Restate your problem 3 different ways (change objective, formalism, agent).
  1. 否定（框架5）：否定3个软约束/隐约束，会产生什么系统？
  2. 双重联想（框架1）：选择一个遥远领域，和你所在领域构建交叉乘积矩阵
  3. 问题重构（框架2）：用3种不同方式重述你的问题（改变目标、形式化表达、主体）

Phase 3: Deepen Promising Leads (30 min)

阶段3：打磨有潜力的方向（30分钟）

  1. Analogical Reasoning (F3): For each promising idea, find a structural analogy and extract predictions.
  2. Abstraction Laddering (F6): Move each idea up (generalize) and down (specialize).
  3. Janusian Thinking (F8): Identify any tensions. Can you synthesize rather than choose?
  1. 类比推理（框架3）：针对每个有潜力的想法，找到结构类比并提取预测
  2. 抽象阶梯（框架6）：把每个想法向上（泛化）和向下（特化）延伸
  3. 两面神思维（框架8）：识别所有张力，能不能综合而不是选边？

Phase 4: Evaluate (15 min)

阶段4：评估（15分钟）

Apply the two-sentence test (from the brainstorm skill):
"[Domain] currently struggles with [problem] because [reason]. We [approach] by [mechanism], which works because [insight]."
Any idea that survives all four phases and passes the two-sentence test is worth pursuing.
对想法应用两句话测试（来自头脑风暴工具）：
「[领域] 当前面临 [问题]，原因是 [根源]。 我们通过 [机制] 采用 [方法]，能够生效的原因是 [洞见]。」
所有通过四个阶段并通过两句话测试的想法都值得推进。

Common Creative Blocks and Unblocking Strategies

常见创意障碍及破解策略

BlockSymptomFramework to Apply
FixationCannot stop thinking about the problem one wayProblem Reformulation (F2) — force a different representation
Tunnel visionAll ideas come from the same subfieldBisociation (F1) or Analogical Reasoning (F3) — import from elsewhere
Self-censoringDismissing ideas as "too weird" before exploringNegation (F5) — weird is the point; evaluate after generating
IncrementalismEvery idea is "+2% on benchmark X"Constraint Manipulation (F4) — change the rules, not the parameters
Analysis paralysisToo many options, cannot commitAdjacent Possible (F7) — what is feasible right now?
False dichotomyStuck choosing between two approachesJanusian Thinking (F8) — seek synthesis, not selection
障碍症状适用框架
思维固化只能用一种方式思考问题问题重构（框架2）——强制换一种表征方式
视野狭窄所有想法都来自同一个细分领域双重联想（框架1）或类比推理（框架3）——从其他领域引入思路
自我审查想法还没探索就被判定为「太奇怪」否定（框架5）——奇怪就是价值，生成阶段结束后再评估
增量主义所有想法都是「在基准测试X上提升2%」约束操作（框架4）——改变规则，而不是调整参数
分析瘫痪选项太多无法决策相邻可能（框架7）——当前什么是可行的？
假二分法卡在两个方案之间选不出两面神思维（框架8）——寻求综合，而不是选择

Usage Instructions for Agents

Agent使用说明

When a researcher asks for help with creative thinking or novel ideation:
  1. Assess the block: What kind of thinking are they stuck in? (See Common Creative Blocks table)
  2. Select 2-3 frameworks based on the block type
  3. Walk through each framework interactively, asking the researcher to supply domain-specific content
  4. Push for structural depth: If an analogy or combination is surface-level, probe deeper
  5. Maintain a running list of all generated ideas, even unusual ones
  6. Apply the two-sentence test to candidates that survive exploration
  7. Hand off to the brainstorm skill for systematic evaluation (diverge → converge → refine)
Key Principles:
  • Generative mode first, evaluative mode second — do not filter prematurely
  • Distant analogies are more valuable than nearby ones, but require more validation
  • The researcher's domain expertise is essential — the agent provides the cognitive scaffolding, not the domain knowledge
  • Encourage the researcher to sit with contradictions rather than resolve them quickly
当研究者寻求创造性思维或新颖构思的帮助时：
  1. 评估障碍类型：他们陷入了哪种思维困境？（参考常见创意障碍表）
  2. 根据障碍类型选择2-3个框架
  3. 交互式引导使用每个框架，请研究者提供领域特定内容
  4. 推动结构性深度：如果类比或组合是表层的，引导深入挖掘
  5. 维护运行中的想法列表，包含所有生成的想法，哪怕是不寻常的
  6. 对通过探索的候选想法应用两句话测试
  7. 转交给头脑风暴工具做系统性评估（发散→收敛→打磨）
核心原则：
  • 先进入生成模式，再进入评估模式——不要过早筛选
  • 遥远类比比邻近类比价值更高，但需要更多验证
  • 研究者的领域专业知识是核心——Agent提供认知脚手架，而非领域知识
  • 鼓励研究者先和矛盾共存，不要急于解决