--- title: '‘truesight (stylometry)’ directory' description: "Bibliography for directory statistics/stylometry/truesight, most recent first: 3 related tags, 44 annotations, & 39 links ()." thumbnail: /doc/statistics/stylometry/truesight/2024-tomekce-figure45-comparisonofvlmtruesightprivacyinferences.png thumbnail-text: '' thumbnail-css: "outline" created: 2023-10-11 modified: 2026-07-07 status: in progress previous: /doc/statistics/stylometry/index next: /doc/statistics/survival-analysis/index confidence: log importance: 0 css-extension: dropcaps-not index: True backlink: False ... # See Also ::: {#see-alsos .directory-indexes .columns} - [Parent ('*`stylometry`{=html}*' tag)](/doc/statistics/stylometry/index "Link to parent directory 'doc/statistics/stylometry/' (ascending)"){#_AsryfEz9 .link-annotated .link-tag .directory-indexes-upwards rel="tag"} - [*`Bing Sydney AI`{=html}*](/doc/ai/nn/transformer/gpt/4/sydney/index "‘Bing Sydney AI’ directory"){#_w5bzQtzI .link-annotated .link-tag .directory-indexes-sideways rel="tag"} - [*`dark knowledge (human)`{=html}*](/doc/psychology/dark-knowledge/index "‘dark knowledge (human)’ directory"){#_xjsaSLQo .link-annotated .link-tag .directory-indexes-sideways rel="tag"} - [*`Decision Transformer`{=html}*](/doc/reinforcement-learning/model/decision-transformer/index "‘Decision Transformer’ directory"){#_e4c3cmDO .link-annotated .link-tag .directory-indexes-sideways rel="tag"} ::: # Gwern {.display-pop-not} ## `“Quantifying Truesight With SAEs”, Gwern 2025`{=html} {#gwern-sae-truesight-section} **[`Quantifying Truesight With SAEs`{=html}](/sae-truesight "'Quantifying Truesight With SAEs', Gwern 2025"){.link-annotated .id-not .include-annotation}** # Links {.display-pop-not} ## `“[Fable Truesight]”, Hua 2026`{=html} {#hua-2026-section} **[`[Fable truesight]`{=html}](https://www.lesswrong.com/posts/NfgukR3HSaRq4WhyW/tim-hua-s-shortform?commentId=fSKR35e4iCv34ppQD "'[Fable truesight]', Hua 2026"){.link-modified-recently .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ::: aux-links-transclude-file ::: collapse **View External Link**: [`https://www.lesswrong.com/posts/NfgukR3HSaRq4WhyW/tim-hua-s-shortform?commentId=fSKR35e4iCv34ppQD`](https://www.lesswrong.com/posts/NfgukR3HSaRq4WhyW/tim-hua-s-shortform?commentId=fSKR35e4iCv34ppQD){.id-not .link-annotated-not .include-content .include-lazy link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"} ::: ::: ## `“[Claude-4.7-Opus Can Truesight Olli Järviniemi]”, Järviniemi 2026`{=html} {#järviniemi-2026-section} **[`[Claude-4.7-opus can truesight Olli Järviniemi]`{=html}](https://www.lesswrong.com/posts/dEER2W3goTsopt48i/olli-jaerviniemi-s-shortform?commentId=4zi6AgGXHkdriRnBa "'[Claude-4.7-opus can truesight Olli Järviniemi]', Järviniemi 2026"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“I Don’t Think We Are close to ‘AI Scientists’; Today’s AI Agents Are Not Designed to Extract Deep Insights from New Observations.”, Lee 2026`{=html} {#lee-2026-section} **[`I don’t think we are close to ‘AI scientists’; today’s AI agents are not designed to extract deep insights from new observations.`{=html}](https://www.understandingai.org/p/i-dont-think-we-are-close-to-ai-scientists "'I don’t think we are close to ‘AI scientists’; today’s AI agents are not designed to extract deep insights from new observations.', Lee 2026"){.link-annotated-partial .id-not .include-annotation}** ## `“A Moderately Well-Known Physicist and I Talked about This a Few Years Ago. He Had…”`{=html} {#_pfD71r7d-section} **[`A moderately well-known physicist and I talked about this a few years ago. He had…`{=html}](https://news.ycombinator.com/item?id=47969329 "A moderately well-known physicist and I talked about this a few years ago. He had…"){.link-annotated-partial .id-not .include-annotation link-icon="hacker-news" link-icon-type="svg" link-icon-color="#f26522"}** ## `“[Claude-4.7-Opus Can Truesight Linch]”, Linch 2026`{=html} {#linch-2026-section} **[`[Claude-4.7-opus can truesight Linch]`{=html}](https://www.lesswrong.com/posts/s58hDHX2GkFDbpGKD/linch-s-shortform?commentId=MoJo8fJztFbXm7NqH "'[Claude-4.7-opus can truesight Linch]', Linch 2026"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“I Can Never Talk to an AI Anonymously Again: AI Only Needs 150 Words to Identify Me. What Does That Mean for You?”, Piper 2026`{=html} {#piper-2026-section} **[`I can never talk to an AI anonymously again: AI only needs 150 words to identify me. What does that mean for you?`{=html}](https://www.theargumentmag.com/p/i-can-never-talk-to-an-ai-anonymously "'I can never talk to an AI anonymously again: AI only needs 150 words to identify me. What does that mean for you?', Piper 2026"){.link-annotated-partial .id-not .include-annotation}** ## `“Claude-4.7-Opus Knows Who You Are”`{=html} {#__PcvQuUG-section} **[`Claude-4.7-opus knows who you are`{=html}](https://www.lesswrong.com/posts/Jkb4CBB7rf4XYP5eb/claude-knows-who-you-are "Claude-4.7-opus knows who you are"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Mapping Synthetic Minds With Janus (Repligate)”, Janus & Ferris 2025`{=html} {#janus-ferris-2025-section} **[`Mapping synthetic minds with Janus (repligate)`{=html}](https://www.goodbyemonkey.com/thegoodtimeline/janus "'Mapping synthetic minds with Janus (repligate)', Janus & Ferris 2025"){.link-annotated-partial .id-not .include-annotation}** ## `“GPT-O3 Used My Saved Pocket Links to Profile Me”, Gross 2025`{=html} {#gross-2025-section} **[`GPT-o3 used my saved Pocket links to profile me`{=html}](https://noperator.dev/posts/o3-pocket-profile/ "'GPT-o3 used my saved Pocket links to profile me', Gross 2025"){.link-annotated-partial .id-not .include-annotation}** ## `“Robustly Improving LLM Fairness in Realistic Settings via Interpretability”, Karvonen & Marks 2025`{=html} {#karvonen-marks-2025-section} **[`Robustly Improving LLM Fairness in Realistic Settings via Interpretability`{=html}](https://arxiv.org/abs/2506.10922 "'Robustly Improving LLM Fairness in Realistic Settings via Interpretability', Karvonen & Marks 2025"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Race and Gender Bias As An Example of Unfaithful Chain-Of-Thought in the Wild”`{=html} {#_iROFA7TE-section} **[`Race and Gender Bias As An Example of Unfaithful Chain-of-Thought in the Wild`{=html}](https://www.lesswrong.com/posts/me7wFrkEtMbkzXGJt/race-and-gender-bias-as-an-example-of-unfaithful-chain-of "Race and Gender Bias As An Example of Unfaithful Chain-of-Thought in the Wild"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“How Does O3 Guess Latitude From Photos?”`{=html} {#_33kWi1WM-section} **[`How Does o3 Guess Latitude From Photos?`{=html}](https://corinwagen.github.io/public/blog/20250527_latitude.html "How Does o3 Guess Latitude From Photos?"){.link-annotated-partial .id-not .include-annotation}** ## `“It’s Really Hard to Make Scheming Evals Look Realistic”`{=html} {#_eCxN05Ts-section} **[`It’s really hard to make scheming evals look realistic`{=html}](https://www.lesswrong.com/posts/TBk2dbWkg2F7dB3jb/it-s-really-hard-to-make-scheming-evals-look-realistic "It’s really hard to make scheming evals look realistic"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Highlights From The Comments On AI Geoguessr”, Alexander 2025`{=html} {#alexander-2025-section} **[`Highlights From The Comments On AI Geoguessr`{=html}](https://www.astralcodexten.com/p/highlights-from-the-comments-on-ai "'Highlights From The Comments On AI Geoguessr', Alexander 2025"){.link-annotated-partial .id-not .include-annotation link-icon="SSC" link-icon-type="text,tri" link-icon-color="#5175c2"}** ## `“Testing AI’s GeoGuessr Genius: Seeing a World in a Grain of Sand”, Alexander 2025`{=html} {#alexander-2025-section} **[`Testing AI’s GeoGuessr Genius: Seeing a world in a grain of sand`{=html}](https://www.astralcodexten.com/p/testing-ais-geoguessr-genius "'Testing AI’s GeoGuessr Genius: Seeing a world in a grain of sand', Alexander 2025"){.link-annotated-partial .id-not .include-annotation link-icon="SSC" link-icon-type="text,tri" link-icon-color="#5175c2"}** ## `“GPT-O3 Beats a Master-Level Geoguessr Player—Even With Fake EXIF Data”, Patterson 2025`{=html} {#patterson-2025-section} **[`GPT-o3 Beats a Master-Level Geoguessr Player—Even with Fake EXIF Data`{=html}](https://sampatt.com/blog/2025-04-28-can-o3-beat-a-geoguessr-master "'GPT-o3 Beats a Master-Level Geoguessr Player—Even with Fake EXIF Data', Patterson 2025"){.link-annotated .id-not .include-annotation}** ## `“Evaluating Precise Geolocation Inference Capabilities of Vision Language Models”, Jay et al 2025`{=html} {#jay-et-al-2025-section} **[`Evaluating Precise Geolocation Inference Capabilities of Vision Language Models`{=html}](https://arxiv.org/abs/2502.14412 "'Evaluating Precise Geolocation Inference Capabilities of Vision Language Models', Jay et al 2025"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Thoughts While Watching Myself Be Automated”, Dynomight 2024`{=html} {#dynomight-2024-section} **[`Thoughts while watching myself be automated`{=html}](https://dynomight.net/automated/ "'Thoughts while watching myself be automated', Dynomight 2024"){.link-annotated .id-not .include-annotation link-icon="D" link-icon-type="text,mono"}** ## `“Investigating the Ability of LLMs to Recognize Their Own Writing”, Ackerman & Panickssery 2024`{=html} {#ackerman-panickssery-2024-section} **[`Investigating the Ability of LLMs to Recognize Their Own Writing`{=html}](https://www.lesswrong.com/posts/ADrTuuus6JsQr5CSi/investigating-the-ability-of-llms-to-recognize-their-own "'Investigating the Ability of LLMs to Recognize Their Own Writing', Ackerman & Panickssery 2024"){.link-annotated .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs”, Laine et al 2024`{=html} {#laine-et-al-2024-section} **[`Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs`{=html}](https://arxiv.org/abs/2407.04694 "'Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs', Laine et al 2024"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Future Events As Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs”, Price et al 2024`{=html} {#price-et-al-2024-2-section} **[`Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs`{=html}](https://arxiv.org/abs/2407.04108 "'Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs', Price et al 2024"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Connecting the Dots: LLMs Can Infer and Verbalize Latent Structure from Disparate Training Data”, Treutlein et al 2024`{=html} {#treutlein-et-al-2024-section} **[`Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data`{=html}](https://arxiv.org/abs/2406.14546 "'Connecting the Dots: LLMs can Infer and Verbalize Latent Structure from Disparate Training Data', Treutlein et al 2024"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Designing a Dashboard for Transparency and Control of Conversational AI”, Chen et al 2024`{=html} {#chen-et-al-2024-4-section} **[`Designing a Dashboard for Transparency and Control of Conversational AI`{=html}](https://arxiv.org/abs/2406.07882 "'Designing a Dashboard for Transparency and Control of Conversational AI', Chen et al 2024"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“LLM Evaluators Recognize and Favor Their Own Generations”, Panickssery et al 2024`{=html} {#panickssery-et-al-2024-section} **[`LLM Evaluators Recognize and Favor Their Own Generations`{=html}](https://arxiv.org/abs/2404.13076 "'LLM Evaluators Recognize and Favor Their Own Generations', Panickssery et al 2024"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Surfing the OCEAN: The Machine Learning Psycholexical Approach 2.0 to Detect Personality Traits in Texts”, Giannini et al 2024`{=html} {#giannini-et-al-2024-section} **[`Surfing the OCEAN: The machine learning psycholexical approach 2.0 to detect personality traits in texts`{=html}](/doc/psychology/personality/2024-giannini.pdf "'Surfing the OCEAN: The machine learning psycholexical approach 2.0 to detect personality traits in texts', Giannini et al 2024"){.link-annotated .id-not .include-annotation link-icon="pdf" link-icon-type="svg" link-icon-color="#f40f02"}** ## `“Beyond Memorization: Violating Privacy Via Inference With Large Language Models”, Staab et al 2023`{=html} {#staab-et-al-2023-section} **[`Beyond Memorization: Violating Privacy Via Inference with Large Language Models`{=html}](https://arxiv.org/abs/2310.07298 "'Beyond Memorization: Violating Privacy Via Inference with Large Language Models', Staab et al 2023"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Taken out of Context: On Measuring Situational Awareness in LLMs”, Berglund et al 2023`{=html} {#berglund-et-al-2023-2-section} **[`Taken out of context: On measuring situational awareness in LLMs`{=html}](https://arxiv.org/abs/2309.00667 "'Taken out of context: On measuring situational awareness in LLMs', Berglund et al 2023"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits”, Jiang et al 2023`{=html} {#jiang-et-al-2023-section} **[`PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits`{=html}](https://arxiv.org/abs/2305.02547 "'PersonaLLM: Investigating the Ability of Large Language Models to Express Personality Traits', Jiang et al 2023"){.link-annotated .id-not .include-annotation link-icon="𝛘" link-icon-type="text" link-icon-color="#b31b1b"}** ## `“Think Before Asking”, Sandberg 2010`{=html} {#sandberg-2010-section} **[`Think Before Asking`{=html}](/doc/fiction/science-fiction/2010-sandberg-eclipsephase-thinkbeforeasking.pdf "'Think Before Asking', Sandberg 2010"){.link-annotated-partial .id-not .include-annotation link-icon="pdf" link-icon-type="svg" link-icon-color="#f40f02"}** ## `“Return Match”, Dick 1967`{=html} {#dick-1967-section} **[`Return Match`{=html}](https://sickmyduck.narod.ru/pkd054-0.html "'Return Match', Dick 1967"){.link-annotated-partial .id-not .include-annotation}** ## `“The Death of Pseudonym”, Wales 2026`{=html} {#_OTGBe5py-section} **[`The Death of Pseudonym`{=html}](https://alexanderwales.substack.com/p/the-death-of-pseudonym "'The Death of Pseudonym', Wales 2026"){.id-not .include-annotation link-icon="substack" link-icon-type="svg" link-icon-color="#ff6719"}** ## `“Truesight”`{=html} {#_aXxNEHi0-section} **[`Truesight`{=html}](https://cyborgism.wiki/hypha/truesight "Truesight"){.link-annotated-partial .id-not .include-annotation}** ## `“Situational Awareness and Out-Of-Context Reasoning § GPT-4-Base Has Non-Zero Longform Performance”, Evans 2026`{=html} {#_0G_nDtqK-section} **[`Situational Awareness and Out-Of-Context Reasoning § GPT-4-base has Non-Zero Longform Performance`{=html}](https://theinsideview.ai/owain#gpt-4-has-non-zero-performance-on-the-longform-task "'Situational Awareness and Out-Of-Context Reasoning § GPT-4-base has Non-Zero Longform Performance', Evans 2026"){.link-annotated-partial .id-not .include-annotation}** ## `“Do Models Continue Misaligned Actions?”`{=html} {#_KrYpgOX2-section} **[`Do Models Continue Misaligned Actions?`{=html}](https://www.lesswrong.com/posts/SawczP2pdCXMrkg2A/do-models-continue-misaligned-actions-2 "Do Models Continue Misaligned Actions?"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Situational Awareness in Large Language Models”`{=html} {#_ZHJDWOwW-section} **[`Situational awareness in Large Language Models`{=html}](https://www.lesswrong.com/posts/TBLv9T7rAzmawehnq/situational-awareness-in-large-language-models "Situational awareness in Large Language Models"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs”`{=html} {#_K2zTkIhs-section} **[`Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs`{=html}](https://www.lesswrong.com/posts/YsCRXZYr5DcJ84XHq/me-myself-and-ai-the-situational-awareness-dataset-sad-for?commentId=qBmEtLM4Tb9bywm8t "Me, Myself, and AI: the Situational Awareness Dataset (SAD) for LLMs"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Language Models Model Us”`{=html} {#__GfDj4tS-section} **[`Language Models Model Us`{=html}](https://www.lesswrong.com/posts/dLg7CyeTE4pqbbcnp/language-models-model-us "Language Models Model Us"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“The Case for More Ambitious Language Model Evals”`{=html} {#_hj1AVBuL-section} **[`The case for more ambitious language model evals`{=html}](https://www.lesswrong.com/posts/doPbyzPgKdjedohud/the-case-for-more-ambitious-language-model-evals?commentId=XZFTx2ek8G8stBKW4 "The case for more ambitious language model evals"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“The Case for More Ambitious Language Model Evals”`{=html} {#_hj1AVBuL-section} **[`The case for more ambitious language model evals`{=html}](https://www.lesswrong.com/posts/doPbyzPgKdjedohud/the-case-for-more-ambitious-language-model-evals?commentId=XZFTx2ek8G8stBKW4 "The case for more ambitious language model evals"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“The Case for More Ambitious Language Model Evals”`{=html} {#_bHpePUnA-section} **[`The case for more ambitious language model evals`{=html}](https://www.lesswrong.com/posts/doPbyzPgKdjedohud/the-case-for-more-ambitious-language-model-evals?commentId=yNcwGQgaJLya8FGah "The case for more ambitious language model evals"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Automated Deanonymization Is Here”, jefftk 2026`{=html} {#_1ld8R4oo-section} **[`Automated Deanonymization is Here`{=html}](https://www.lesswrong.com/posts/dqc8WCQuHaDGBmti4/automated-deanonymization-is-here "'Automated Deanonymization is Here', jefftk 2026"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Early Situational Awareness and Its Implications, a Story”`{=html} {#_L0DnlFho-section} **[`Early situational awareness and its implications, a story`{=html}](https://www.lesswrong.com/posts/tJzdzGdTGrqFf9ekw/early-situational-awareness-and-its-implications-a-story "Early situational awareness and its implications, a story"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“Do Models Know When They Are Being Evaluated?”`{=html} {#_--i2V67t-section} **[`Do models know when they are being evaluated?`{=html}](https://www.lesswrong.com/posts/yTameAzCdycci68sk/do-models-know-when-they-are-being-evaluated "Do models know when they are being evaluated?"){.link-annotated-partial .id-not .include-annotation link-icon="LW" link-icon-type="text" link-icon-color="#7faf83"}** ## `“One Shockingly Impressive Capability of GPT-4.5 [Photo Geolocation]”`{=html} {#_qepcgsYN-section} **[`One shockingly impressive capability of GPT-4.5 [photo geolocation]`{=html}](https://www.linkedin.com/posts/peter-gostev_one-shockingly-impressive-capability-of-gpt-activity-7301990052614574080-PeBv "One shockingly impressive capability of GPT-4.5 [photo geolocation]"){.link-annotated-partial .id-not .include-annotation}** ## Sort By Magic ::: {demo-type="sort-by-magic-preface"} Annotations sorted by machine learning into [inferred \'tags\'](/design#future-tag-features){#gwern-design--future-tag-features .link-annotated-partial}. This provides an alternative way to browse: instead of by *date* order, one can browse in *topic* order. The \'sorted\' list has been automatically clustered into multiple sections & auto-labeled for easier browsing. Beginning with the newest annotation, it uses the embedding of each annotation to attempt to create a list of nearest-neighbor annotations, creating a progression of topics. For more details, see the link. ::: ### `temporal-vulnerabilities` {.collapse title="Machine-generated tag name for the following cluster of links."} [\[see previous entry\]](#price-et-al-2024-2-section){.include .include-even-when-collapsed} ### `geoguessr ai-game exif-manipulation automation commentary automation-analytics geoguessr-analytics` {.collapse title="Machine-generated tag name for the following cluster of links."} [\[see previous entry\]](#patterson-2025-section){.include .include-even-when-collapsed} [\[see previous entry\]](#alexander-2025-section){.include .include-even-when-collapsed} [\[see previous entry\]](#dynomight-2024-section){.include .include-even-when-collapsed} ### `llm-evaluation llm-inference privacy-violation situational-awareness interpretability-dashboard transparency-control` {.collapse title="Machine-generated tag name for the following cluster of links."} [\[see previous entry\]](#ackerman-panickssery-2024-section){.include .include-even-when-collapsed} [\[see previous entry\]](#panickssery-et-al-2024-section){.include .include-even-when-collapsed} [\[see previous entry\]](#karvonen-marks-2025-section){.include .include-even-when-collapsed} [\[see previous entry\]](#berglund-et-al-2023-2-section){.include .include-even-when-collapsed} [\[see previous entry\]](#treutlein-et-al-2024-section){.include .include-even-when-collapsed} [\[see previous entry\]](#laine-et-al-2024-section){.include .include-even-when-collapsed} [\[see previous entry\]](#chen-et-al-2024-4-section){.include .include-even-when-collapsed} [\[see previous entry\]](#staab-et-al-2023-section){.include .include-even-when-collapsed} [\[see previous entry\]](#jay-et-al-2025-section){.include .include-even-when-collapsed} ### `personality-detection` {.collapse title="Machine-generated tag name for the following cluster of links."} [\[see previous entry\]](#jiang-et-al-2023-section){.include .include-even-when-collapsed} [\[see previous entry\]](#giannini-et-al-2024-section){.include .include-even-when-collapsed} # Miscellaneous - **[`https://www.washingtonpost.com/opinions/interactive/2026/04/26/artificial-intelligence-could-kill-anonymity-online/`{=html}](https://www.washingtonpost.com/opinions/interactive/2026/04/26/artificial-intelligence-could-kill-anonymity-online/){.id-not .include-annotation link-icon="washington-post" link-icon-type="svg"}** - **[`/doc/statistics/stylometry/truesight/2024-tomekce-figure45-comparisonofvlmtruesightprivacyinferences.png`{=html}](/doc/statistics/stylometry/truesight/2024-tomekce-figure45-comparisonofvlmtruesightprivacyinferences.png){.link-annotated-partial .id-not .include-annotation link-icon="image" link-icon-type="svg"}** - **[`https://x.com/patio11/status/1917395208368529723`{=html}](https://x.com/patio11/status/1917395208368529723 "McKenzie 2023"){.id-not .include-annotation link-icon="pt11" link-icon-type="text,quad,mono" link-icon-color="#3498db"}** ::: aux-links-transclude-file [`https://x.com/patio11/status/1917395208368529723`](https://x.com/patio11/status/1917395208368529723 "McKenzie 2023"){.id-not .include-content} ::: - **[`https://x.com/Afinetheorem/status/1912966709654560930`{=html}](https://x.com/Afinetheorem/status/1912966709654560930 "Afinetheorem 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/Afinetheorem/status/1912966709654560930`](https://x.com/Afinetheorem/status/1912966709654560930 "Afinetheorem 2026"){.id-not .include-content} ::: - **[`https://x.com/Altimor/status/1873872492412231776`{=html}](https://x.com/Altimor/status/1873872492412231776 "Crivello 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/Altimor/status/1873872492412231776`](https://x.com/Altimor/status/1873872492412231776 "Crivello 2026"){.id-not .include-content} ::: - **[`https://x.com/AndyAyrey/status/1810869652484149486`{=html}](https://x.com/AndyAyrey/status/1810869652484149486 "AndyAyrey 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/AndyAyrey/status/1810869652484149486`](https://x.com/AndyAyrey/status/1810869652484149486 "AndyAyrey 2026"){.id-not .include-content} ::: - **[`https://x.com/AstronautSwing/status/1819902419272171583`{=html}](https://x.com/AstronautSwing/status/1819902419272171583 "AstronautSwing 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/AstronautSwing/status/1819902419272171583`](https://x.com/AstronautSwing/status/1819902419272171583 "AstronautSwing 2026"){.id-not .include-content} ::: - **[`https://x.com/DimitrisPapail/status/1925385259069190525`{=html}](https://x.com/DimitrisPapail/status/1925385259069190525 "Papail 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/DimitrisPapail/status/1925385259069190525`](https://x.com/DimitrisPapail/status/1925385259069190525 "Papail 2026"){.id-not .include-content} ::: - **[`https://x.com/KelseyTuoc/status/1917344695988150670`{=html}](https://x.com/KelseyTuoc/status/1917344695988150670 "Piper 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/KelseyTuoc/status/1917344695988150670`](https://x.com/KelseyTuoc/status/1917344695988150670 "Piper 2026"){.id-not .include-content} ::: - **[`https://x.com/KelseyTuoc/status/1918045579835478453`{=html}](https://x.com/KelseyTuoc/status/1918045579835478453 "Piper 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/KelseyTuoc/status/1918045579835478453`](https://x.com/KelseyTuoc/status/1918045579835478453 "Piper 2026"){.id-not .include-content} ::: - **[`https://x.com/KelseyTuoc/status/2044962433379491971`{=html}](https://x.com/KelseyTuoc/status/2044962433379491971 "Piper 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/KelseyTuoc/status/2044962433379491971`](https://x.com/KelseyTuoc/status/2044962433379491971 "Piper 2026"){.id-not .include-content} ::: - **[`https://x.com/QiaochuYuan/status/1917371306905067837`{=html}](https://x.com/QiaochuYuan/status/1917371306905067837 "Yuan 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/QiaochuYuan/status/1917371306905067837`](https://x.com/QiaochuYuan/status/1917371306905067837 "Yuan 2026"){.id-not .include-content} ::: - **[`https://x.com/QiaochuYuan/status/2046646692725006708`{=html}](https://x.com/QiaochuYuan/status/2046646692725006708 "Yuan 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/QiaochuYuan/status/2046646692725006708`](https://x.com/QiaochuYuan/status/2046646692725006708 "Yuan 2026"){.id-not .include-content} ::: - **[`https://x.com/RichardHanania/status/2048918204366684530`{=html}](https://x.com/RichardHanania/status/2048918204366684530 "Hanania 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/RichardHanania/status/2048918204366684530`](https://x.com/RichardHanania/status/2048918204366684530 "Hanania 2026"){.id-not .include-content} ::: - **[`https://x.com/Sauers_/status/1850678934997754127`{=html}](https://x.com/Sauers_/status/1850678934997754127 "Sauers 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/Sauers_/status/1850678934997754127`](https://x.com/Sauers_/status/1850678934997754127 "Sauers 2026"){.id-not .include-content} ::: - **[`https://x.com/Sauers_/status/2069158522676220169`{=html}](https://x.com/Sauers_/status/2069158522676220169 "Sauers 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/Sauers_/status/2069158522676220169`](https://x.com/Sauers_/status/2069158522676220169 "Sauers 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/StefanFSchubert/status/1873513175670501606`{=html}](https://x.com/StefanFSchubert/status/1873513175670501606 "StefanFSchubert 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/StefanFSchubert/status/1873513175670501606`](https://x.com/StefanFSchubert/status/1873513175670501606 "StefanFSchubert 2026"){.id-not .include-content} ::: - **[`https://x.com/akoustov/status/2045205486840984026`{=html}](https://x.com/akoustov/status/2045205486840984026 "akoustov 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/akoustov/status/2045205486840984026`](https://x.com/akoustov/status/2045205486840984026 "akoustov 2026"){.id-not .include-content} ::: - **[`https://x.com/aryaman2020/status/2064975916397416461`{=html}](https://x.com/aryaman2020/status/2064975916397416461 "aryaman2020 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/aryaman2020/status/2064975916397416461`](https://x.com/aryaman2020/status/2064975916397416461 "aryaman2020 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/avt_im/status/2046071624433016866`{=html}](https://x.com/avt_im/status/2046071624433016866 "avt_im 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/avt_im/status/2046071624433016866`](https://x.com/avt_im/status/2046071624433016866 "avt_im 2026"){.id-not .include-content} ::: - **[`https://x.com/doomslide/status/1830149217521672373`{=html}](https://x.com/doomslide/status/1830149217521672373 "doomslide 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/doomslide/status/1830149217521672373`](https://x.com/doomslide/status/1830149217521672373 "doomslide 2026"){.id-not .include-content} ::: - **[`https://x.com/ericneyman/status/2045173161139089656`{=html}](https://x.com/ericneyman/status/2045173161139089656 "ericneyman 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/ericneyman/status/2045173161139089656`](https://x.com/ericneyman/status/2045173161139089656 "ericneyman 2026"){.id-not .include-content} ::: - **[`https://x.com/gdere/status/2065469412275073198`{=html}](https://x.com/gdere/status/2065469412275073198 "gdere 2026"){.link-modified-recently .link-annotated-partial .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** - **[`https://x.com/gleech/status/2064687582726824295`{=html}](https://x.com/gleech/status/2064687582726824295 "Leech 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/gleech/status/2064687582726824295`](https://x.com/gleech/status/2064687582726824295 "Leech 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/jd_pressman/status/1808398225260569016`{=html}](https://x.com/jd_pressman/status/1808398225260569016 "jd_pressman 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/jd_pressman/status/1808398225260569016`](https://x.com/jd_pressman/status/1808398225260569016 "jd_pressman 2026"){.id-not .include-content} ::: - **[`https://x.com/lionellevine/status/2045002945855398335`{=html}](https://x.com/lionellevine/status/2045002945855398335 "lionellevine 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/lionellevine/status/2045002945855398335`](https://x.com/lionellevine/status/2045002945855398335 "lionellevine 2026"){.id-not .include-content} ::: - **[`https://x.com/mehran__jalali/status/1873605917301973285`{=html}](https://x.com/mehran__jalali/status/1873605917301973285 "Jalali 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/mehran__jalali/status/1873605917301973285`](https://x.com/mehran__jalali/status/1873605917301973285 "Jalali 2026"){.id-not .include-content} ::: - **[`https://x.com/qwrk8126/status/1873912766588920006`{=html}](https://x.com/qwrk8126/status/1873912766588920006 "qwrk8126 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/qwrk8126/status/1873912766588920006`](https://x.com/qwrk8126/status/1873912766588920006 "qwrk8126 2026"){.id-not .include-content} ::: - **[`https://x.com/repligate/status/1806993408818299166`{=html}](https://x.com/repligate/status/1806993408818299166 "Janus 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/repligate/status/1806993408818299166`](https://x.com/repligate/status/1806993408818299166 "Janus 2026"){.id-not .include-content} ::: - **[`https://x.com/repligate/status/1808396202146136099`{=html}](https://x.com/repligate/status/1808396202146136099 "Janus 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/repligate/status/1808396202146136099`](https://x.com/repligate/status/1808396202146136099 "Janus 2026"){.id-not .include-content} ::: - **[`https://x.com/repligate/status/1828266415851208803`{=html}](https://x.com/repligate/status/1828266415851208803 "Janus 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/repligate/status/1828266415851208803`](https://x.com/repligate/status/1828266415851208803 "Janus 2026"){.id-not .include-content} ::: - **[`https://x.com/sharifshameem/status/1851059380730613776`{=html}](https://x.com/sharifshameem/status/1851059380730613776 "Shameem 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/sharifshameem/status/1851059380730613776`](https://x.com/sharifshameem/status/1851059380730613776 "Shameem 2026"){.id-not .include-content} ::: - **[`https://x.com/tracewoodgrains/status/2064426141838942530`{=html}](https://x.com/tracewoodgrains/status/2064426141838942530 "tracewoodgrains 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/tracewoodgrains/status/2064426141838942530`](https://x.com/tracewoodgrains/status/2064426141838942530 "tracewoodgrains 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/tracewoodgrains/status/2074442842823618747`{=html}](https://x.com/tracewoodgrains/status/2074442842823618747 "tracewoodgrains 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/tracewoodgrains/status/2074442842823618747`](https://x.com/tracewoodgrains/status/2074442842823618747 "tracewoodgrains 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/venturetwins/status/1822682396090937538`{=html}](https://x.com/venturetwins/status/1822682396090937538 "Moore 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/venturetwins/status/1822682396090937538`](https://x.com/venturetwins/status/1822682396090937538 "Moore 2026"){.id-not .include-content} ::: - **[`https://x.com/voooooogel/status/1830797676243492947`{=html}](https://x.com/voooooogel/status/1830797676243492947 "Luoma 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/voooooogel/status/1830797676243492947`](https://x.com/voooooogel/status/1830797676243492947 "Luoma 2026"){.id-not .include-content} ::: - **[`https://x.com/xuenay/status/1873502684076228821`{=html}](https://x.com/xuenay/status/1873502684076228821 "xuenay 2026"){.id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/xuenay/status/1873502684076228821`](https://x.com/xuenay/status/1873502684076228821 "xuenay 2026"){.id-not .include-content} ::: - **[`https://x.com/zetalyrae/status/2064892203881955623`{=html}](https://x.com/zetalyrae/status/2064892203881955623 "Borretti 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/zetalyrae/status/2064892203881955623`](https://x.com/zetalyrae/status/2064892203881955623 "Borretti 2026"){.link-modified-recently .id-not .include-content} ::: - **[`https://x.com/zetalyrae/status/2068181110727577827`{=html}](https://x.com/zetalyrae/status/2068181110727577827 "Borretti 2026"){.link-modified-recently .id-not .include-annotation link-icon="twitter" link-icon-type="svg" link-icon-color="#1da1f2"}** ::: aux-links-transclude-file [`https://x.com/zetalyrae/status/2068181110727577827`](https://x.com/zetalyrae/status/2068181110727577827 "Borretti 2026"){.link-modified-recently .id-not .include-content} ::: # Bibliography {#link-bibliography-section} 1. `https://arxiv.org/abs/2506.10922`: [`“Robustly Improving LLM Fairness in Realistic Settings via Interpretability”`{=html}](https://arxiv.org/abs/2506.10922 "'Robustly Improving LLM Fairness in Realistic Settings via Interpretability', Karvonen & Marks 2025"){#karvonen-marks-2025 .link-annotated}, [Adam Karvonen, [Samuel Marks](https://scholar.google.com/citations?user=fW7yK10AAAAJ "Samuel Marks"){.id-not}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2506.10922.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2506.10922"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 2. `https://dynomight.net/automated/`: [`“Thoughts While Watching Myself Be Automated”`{=html}](https://dynomight.net/automated/ "'Thoughts while watching myself be automated', Dynomight 2024"){#dynomight-2024 .link-annotated}, [Dynomight]{.author .cite-author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Fdynomight.net%252Fautomated%252F.html "Directory-tag link-bibliography for link https://dynomight.net/automated/"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 3. `https://www.lesswrong.com/posts/ADrTuuus6JsQr5CSi/investigating-the-ability-of-llms-to-recognize-their-own`: [`“Investigating the Ability of LLMs to Recognize Their Own Writing”`{=html}](https://www.lesswrong.com/posts/ADrTuuus6JsQr5CSi/investigating-the-ability-of-llms-to-recognize-their-own "'Investigating the Ability of LLMs to Recognize Their Own Writing', Ackerman & Panickssery 2024"){#ackerman-panickssery-2024 .link-annotated}, [Christopher Ackerman, [Nina Panickssery](https://ninapanickssery.com/research){.link-annotated-partial .id-not}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Fwww.lesswrong.com%252Fposts%252FADrTuuus6JsQr5CSi%252Finvestigating-the-ability-of-llms-to-recognize-their-own.html "Directory-tag link-bibliography for link https://www.lesswrong.com/posts/ADrTuuus6JsQr5CSi/investigating-the-ability-of-llms-to-recognize-their-own"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 4. `https://arxiv.org/abs/2407.04694`: [`“Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs”`{=html}](https://arxiv.org/abs/2407.04694 "'Me, Myself, and AI: The Situational Awareness Dataset (SAD) for LLMs', Laine et al 2024"){#laine-et-al-2024 .link-annotated}, [Rudolf Laine, Bilal Chughtai, Jan Betley[, Kaivalya Hariharan, Jeremy Scheurer, Mikita Balesni, Marius Hobbhahn, Alexander Meinke, [Owain Evans](https://owainevans.github.io/ "Owain Evans, AI Alignment researcher"){.link-annotated-partial .id-not}]{.collapse}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2407.04694.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2407.04694"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 5. `https://arxiv.org/abs/2407.04108`: [`“Future Events As Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs”`{=html}](https://arxiv.org/abs/2407.04108 "'Future Events as Backdoor Triggers: Investigating Temporal Vulnerabilities in LLMs', Price et al 2024"){#price-et-al-2024-2 .link-annotated}, [Sara Price, Arjun Panickssery, [Samuel R. Bowman](https://cims.nyu.edu/~sbowman/ "Sam Bowman"){.link-annotated-partial .id-not}, Asa Cooper Stickland]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2407.04108.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2407.04108"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 6. `https://arxiv.org/abs/2406.07882`: [`“Designing a Dashboard for Transparency and Control of Conversational AI”`{=html}](https://arxiv.org/abs/2406.07882 "'Designing a Dashboard for Transparency and Control of Conversational AI', Chen et al 2024"){#chen-et-al-2024-4 .link-annotated}, [Yida Chen, Aoyu Wu, Trevor DePodesta[, Catherine Yeh, Kenneth Li, Nicholas Castillo Marin, Oam Patel, Jan Riecke, Shivam Raval, Olivia Seow, [Martin M. Wattenberg](https://en.wikipedia.org/wiki/Martin_M._Wattenberg "Martin M. Wattenberg"){.id-not}, [Fernanda Viégas](https://en.wikipedia.org/wiki/Fernanda_Vi%C3%A9gas "Fernanda Viégas"){.id-not}]{.collapse}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2406.07882.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2406.07882"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 7. `https://arxiv.org/abs/2404.13076`: [`“LLM Evaluators Recognize and Favor Their Own Generations”`{=html}](https://arxiv.org/abs/2404.13076 "'LLM Evaluators Recognize and Favor Their Own Generations', Panickssery et al 2024"){#panickssery-et-al-2024 .link-annotated}, [Arjun Panickssery, [Samuel R. Bowman](https://cims.nyu.edu/~sbowman/ "Sam Bowman"){.link-annotated-partial .id-not}, [Shi Feng](https://ihsgnef.github.io/ "Shi Feng"){.id-not}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2404.13076.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2404.13076"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block} 8. `https://arxiv.org/abs/2309.00667`: [`“Taken out of Context: On Measuring Situational Awareness in LLMs”`{=html}](https://arxiv.org/abs/2309.00667 "'Taken out of context: On measuring situational awareness in LLMs', Berglund et al 2023"){#berglund-et-al-2023-2 .link-annotated}, [Lukas Berglund, Asa Cooper Stickland, Mikita Balesni[, Max Kaufmann, Meg Tong, Tomasz Korbak, [Daniel Kokotajlo](https://en.wikipedia.org/wiki/Daniel_Kokotajlo "Daniel Kokotajlo"){.id-not}, [Owain Evans](https://owainevans.github.io/ "Owain Evans, AI Alignment researcher"){.link-annotated-partial .id-not}]{.collapse}]{.author} [[link-bibliography](/metadata/annotation/link-bibliography/https%253A%252F%252Farxiv.org%252Fabs%252F2309.00667.html "Directory-tag link-bibliography for link https://arxiv.org/abs/2309.00667"){.id-not .include-even-when-collapsed}]{.collapse .tag-index-link-bibliography-block}