{"site":"https://introspection.infinite.fun","nodes":[{"id":"atkinson2026-identifying-introspection","title":"Identifying Introspection From the Inside","year":2026,"tier":"seed","url":"https://introspection.infinite.fun/papers/atkinson2026-identifying-introspection"},{"id":"binder2024-looking-inward","title":"Looking Inward: Language Models Can Learn About Themselves by Introspection","year":2024,"tier":"core","url":"https://introspection.infinite.fun/papers/binder2024-looking-inward"},{"id":"sherburn2024-explain-classification-behavior","title":"Can Language Models Explain Their Own Classification Behavior?","year":2024,"tier":"core","url":"https://introspection.infinite.fun/papers/sherburn2024-explain-classification-behavior"},{"id":"betley2025-tell-me-about-yourself","title":"Tell me about yourself: LLMs are aware of their learned behaviors","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/betley2025-tell-me-about-yourself"},{"id":"comsa2025-speak-of-introspection","title":"Does It Make Sense to Speak of Introspection in Large Language Models?","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/comsa2025-speak-of-introspection"},{"id":"li2025-explain-own-computations","title":"Training Language Models to Explain Their Own Computations","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/li2025-explain-own-computations"},{"id":"lindsey2025-emergent-introspective-awareness","title":"Emergent Introspective Awareness in Large Language Models","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/lindsey2025-emergent-introspective-awareness"},{"id":"morris2025-causal-bypassing","title":"Tests of LLM introspection need to rule out causal bypassing","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/morris2025-causal-bypassing"},{"id":"plunkett2025-self-interpretability","title":"Self-Interpretability: LLMs Can Describe Complex Internal Processes that Drive Their Decisions, and Improve with Training","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/plunkett2025-self-interpretability"},{"id":"song2025-fail-to-introspect","title":"Language Models Fail to Introspect About Their Knowledge of Language","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/song2025-fail-to-introspect"},{"id":"song2025-privileged-self-access","title":"Privileged Self-Access Matters for Introspection in AI","year":2025,"tier":"core","url":"https://introspection.infinite.fun/papers/song2025-privileged-self-access"},{"id":"hahami2026-detecting-the-disturbance","title":"Detecting the Disturbance: A Nuanced View of Introspective Abilities in LLMs","year":2026,"tier":"core","url":"https://introspection.infinite.fun/papers/hahami2026-detecting-the-disturbance"},{"id":"pearson-vogel2026-latent-introspection","title":"Latent Introspection: Models Can Detect Prior Concept Injections","year":2026,"tier":"core","url":"https://introspection.infinite.fun/papers/pearson-vogel2026-latent-introspection"},{"id":"berglund2023-taken-out-of-context","title":"Taken out of context: On measuring situational awareness in LLMs","year":2023,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/berglund2023-taken-out-of-context"},{"id":"treutlein2024-connecting-the-dots","title":"Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data","year":2024,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/treutlein2024-connecting-the-dots"},{"id":"bai2025-explicitly-unbiased","title":"Explicitly unbiased large language models still form biased associations","year":2025,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/bai2025-explicitly-unbiased"},{"id":"cywinski2025-eliciting-secret-knowledge","title":"Eliciting Secret Knowledge from Language Models","year":2025,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/cywinski2025-eliciting-secret-knowledge"},{"id":"lindsey2025-biology-of-llm","title":"On the Biology of a Large Language Model","year":2025,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/lindsey2025-biology-of-llm"},{"id":"wang2025-mechanistic-oocr","title":"Simple Mechanistic Explanations for Out-Of-Context Reasoning","year":2025,"tier":"adjacent","url":"https://introspection.infinite.fun/papers/wang2025-mechanistic-oocr"}],"edges":[{"source":"atkinson2026-identifying-introspection","target":"binder2024-looking-inward"},{"source":"atkinson2026-identifying-introspection","target":"sherburn2024-explain-classification-behavior"},{"source":"atkinson2026-identifying-introspection","target":"betley2025-tell-me-about-yourself"},{"source":"atkinson2026-identifying-introspection","target":"comsa2025-speak-of-introspection"},{"source":"atkinson2026-identifying-introspection","target":"li2025-explain-own-computations"},{"source":"atkinson2026-identifying-introspection","target":"lindsey2025-emergent-introspective-awareness"},{"source":"atkinson2026-identifying-introspection","target":"morris2025-causal-bypassing"},{"source":"atkinson2026-identifying-introspection","target":"plunkett2025-self-interpretability"},{"source":"atkinson2026-identifying-introspection","target":"song2025-fail-to-introspect"},{"source":"atkinson2026-identifying-introspection","target":"song2025-privileged-self-access"},{"source":"atkinson2026-identifying-introspection","target":"hahami2026-detecting-the-disturbance"},{"source":"atkinson2026-identifying-introspection","target":"pearson-vogel2026-latent-introspection"},{"source":"atkinson2026-identifying-introspection","target":"berglund2023-taken-out-of-context"},{"source":"atkinson2026-identifying-introspection","target":"treutlein2024-connecting-the-dots"},{"source":"atkinson2026-identifying-introspection","target":"bai2025-explicitly-unbiased"},{"source":"atkinson2026-identifying-introspection","target":"cywinski2025-eliciting-secret-knowledge"},{"source":"atkinson2026-identifying-introspection","target":"lindsey2025-biology-of-llm"},{"source":"atkinson2026-identifying-introspection","target":"wang2025-mechanistic-oocr"},{"source":"binder2024-looking-inward","target":"berglund2023-taken-out-of-context"},{"source":"binder2024-looking-inward","target":"treutlein2024-connecting-the-dots"},{"source":"betley2025-tell-me-about-yourself","target":"binder2024-looking-inward"},{"source":"betley2025-tell-me-about-yourself","target":"berglund2023-taken-out-of-context"},{"source":"betley2025-tell-me-about-yourself","target":"treutlein2024-connecting-the-dots"},{"source":"comsa2025-speak-of-introspection","target":"binder2024-looking-inward"},{"source":"li2025-explain-own-computations","target":"binder2024-looking-inward"},{"source":"li2025-explain-own-computations","target":"comsa2025-speak-of-introspection"},{"source":"li2025-explain-own-computations","target":"plunkett2025-self-interpretability"},{"source":"li2025-explain-own-computations","target":"song2025-fail-to-introspect"},{"source":"li2025-explain-own-computations","target":"song2025-privileged-self-access"},{"source":"li2025-explain-own-computations","target":"treutlein2024-connecting-the-dots"},{"source":"plunkett2025-self-interpretability","target":"binder2024-looking-inward"},{"source":"plunkett2025-self-interpretability","target":"betley2025-tell-me-about-yourself"},{"source":"song2025-fail-to-introspect","target":"binder2024-looking-inward"},{"source":"song2025-fail-to-introspect","target":"betley2025-tell-me-about-yourself"},{"source":"song2025-privileged-self-access","target":"binder2024-looking-inward"},{"source":"song2025-privileged-self-access","target":"betley2025-tell-me-about-yourself"},{"source":"song2025-privileged-self-access","target":"comsa2025-speak-of-introspection"},{"source":"hahami2026-detecting-the-disturbance","target":"binder2024-looking-inward"},{"source":"hahami2026-detecting-the-disturbance","target":"lindsey2025-emergent-introspective-awareness"},{"source":"hahami2026-detecting-the-disturbance","target":"morris2025-causal-bypassing"},{"source":"pearson-vogel2026-latent-introspection","target":"binder2024-looking-inward"},{"source":"pearson-vogel2026-latent-introspection","target":"lindsey2025-emergent-introspective-awareness"},{"source":"pearson-vogel2026-latent-introspection","target":"song2025-privileged-self-access"},{"source":"treutlein2024-connecting-the-dots","target":"berglund2023-taken-out-of-context"},{"source":"cywinski2025-eliciting-secret-knowledge","target":"betley2025-tell-me-about-yourself"},{"source":"wang2025-mechanistic-oocr","target":"betley2025-tell-me-about-yourself"},{"source":"wang2025-mechanistic-oocr","target":"berglund2023-taken-out-of-context"},{"source":"wang2025-mechanistic-oocr","target":"treutlein2024-connecting-the-dots"}]}