newsGoogle News — LLMTrust 62 · AggregatorPublished 1mo agoLive · 1mo ago

Understanding large language models demands distinguishing human projection from machine cognition - Nature

Understanding large language models demands distinguishing human projection from machine cognition Nature

Why these links exist

Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.

PossiblePossibly related (embedding) · 60%Understanding Large Language Models →
PossiblePossibly related (embedding) · 56%Conversable Complexity: Agentic LLM Collectives as Interpretable Substrates →
PossiblePossibly related (embedding) · 55%Furyton/awesome-language-model-analysis →
PossiblePossibly related (embedding) · 55%Representational Depth of Evaluation Awareness Shifts With Scale in Open-Weight Language Models →
PossiblePossibly related (embedding) · 54%Language Models as Measurement Apparatus for Culture →
PossiblePossibly related (embedding) · 48%On the Threat Model of Weird Generalization and Emergent Misalignment →
PossiblePossibly related (embedding) · 52%A Formal Limitation on Learning Human Language From Textual Corpora →
PossiblePossibly related (embedding) · 48%Do Vision-Language Models Agree on the Affective Qualities of Shape? A Cross-Model Audit for Generative Design Interfaces →

Covers

paperUnderstanding Large Language Models paperConversable Complexity: Agentic LLM Collectives as Interpretable Substrates repoFuryton/awesome-language-model-analysis paperRepresentational Depth of Evaluation Awareness Shifts With Scale in Open-Weight Language Models paperLanguage Models as Measurement Apparatus for Culture

Covers (incoming)

paperOn the Threat Model of Weird Generalization and Emergent Misalignment paperA Formal Limitation on Learning Human Language From Textual Corpora paperDo Vision-Language Models Agree on the Affective Qualities of Shape? A Cross-Model Audit for Generative Design Interfaces paperThink-Probe-Respond: Improving Large Language Models as Judges of Research Idea Novelty paperDistinct dynamics of conceptual and referential disruptions in human reading and large language model processing paperWhen Persona Attributes Improve Population Alignment in Large Language Models paperLinear representations of grammaticality in neural language models paperHoloGeo: Mitigating Landmark Bias in Geo-localization via Evidence-Driven Reasoning paperBefore the Action: Benchmarking LLMs on Prospective Hypothesis Discovery paperHow Much Human Label Variation Does Formal Semantic Structure Explain?: Group-Level Effects and Item-Level Ceilings in NLI paperIt's Not What You Say, It's How You Say It: Evaluating LLM Responses to Expressions of Belief paperThe Two-Process Theory of Machine Self-Report paperGotta Catch them all: the modes of Sycophancy papersurprisal is Not a Theory paperExposure is Optional: Learning Unlike Coordination in Language Models paperThe Maskability Index: Predicting Task-Objective Alignment in Pretrained Language Models paperEmergent Misalignment Recruits a Pre-existing Persona Subspace paperSurprisal Theory is Tautological (without Rational Grounding)paperAre You Sure You're Sure? On the Impact of Instruction Tuning on Confidence and Lexical Diversity paperSeeing Red, Thinking Bad: Color Bias in Vision Language Models paperAnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs paperLocal and Global Regimes of Geometric Complexity in Language Model Representations paperInterpretable Humans, Alien LLMs: Expert Analysis of Latent Structures in Assessment Responses paperDo Large Language Models Hallucinate Electric Fata Morganas?paperReWEIGH the Evidence: Calibrating Token-Level Ordinal Visual Evidence to Mitigate Hallucinations in Large Vision-Language Models paperWhen the Feature Pool Goes Algorithmic: Extending Mufwene's Ecology of Language Evolution to LLM-Mediated Exposure

Related across the graph

paperHow Much Human Label Variation Does Formal Semantic Structure Explain?: Group-Level Effects and Item-Level Ceilings in NLI paperEmergent Misalignment Recruits a Pre-existing Persona Subspace paperBefore the Action: Benchmarking LLMs on Prospective Hypothesis Discovery paperReWEIGH the Evidence: Calibrating Token-Level Ordinal Visual Evidence to Mitigate Hallucinations in Large Vision-Language Models paperSurprisal Theory is Tautological (without Rational Grounding)paperSeeing Red, Thinking Bad: Color Bias in Vision Language Models paperDo Vision-Language Models Agree on the Affective Qualities of Shape? A Cross-Model Audit for Generative Design Interfaces paperExposure is Optional: Learning Unlike Coordination in Language Models paperIt's Not What You Say, It's How You Say It: Evaluating LLM Responses to Expressions of Belief paperHoloGeo: Mitigating Landmark Bias in Geo-localization via Evidence-Driven Reasoning paperInterpretable Humans, Alien LLMs: Expert Analysis of Latent Structures in Assessment Responses paperOn the Threat Model of Weird Generalization and Emergent Misalignment paperRepresentational Depth of Evaluation Awareness Shifts With Scale in Open-Weight Language Models paperWhen Persona Attributes Improve Population Alignment in Large Language Models paperAre You Sure You're Sure? On the Impact of Instruction Tuning on Confidence and Lexical Diversity paperThe Two-Process Theory of Machine Self-Report paperLocal and Global Regimes of Geometric Complexity in Language Model Representations paperA Formal Limitation on Learning Human Language From Textual Corpora paperAnchorBench: A Multi-Pathway Benchmark for the Anchoring Effect in LLMs paperLanguage Models as Measurement Apparatus for Culture paperGotta Catch them all: the modes of Sycophancy paperUnderstanding Large Language Models paperConversable Complexity: Agentic LLM Collectives as Interpretable Substrates paperDo Large Language Models Hallucinate Electric Fata Morganas?paperThe Maskability Index: Predicting Task-Objective Alignment in Pretrained Language Models papersurprisal is Not a Theory paperLinear representations of grammaticality in neural language models paperThink-Probe-Respond: Improving Large Language Models as Judges of Research Idea Novelty paperWhen the Feature Pool Goes Algorithmic: Extending Mufwene's Ecology of Language Evolution to LLM-Mediated Exposure repoFuryton/awesome-language-model-analysis paperDistinct dynamics of conceptual and referential disruptions in human reading and large language model processing