glossary termAngestromTrust 60Published 2mo agoLive · 3mo ago
Alignment
Making a model's behavior match human intent and values.
Making a model's behavior match human intent and values. Making a model's behavior match human intent and values.
Why these links exist
Every edge carries a method, confidence, and the source snippet that justified it — so bad links are debuggable.
- LinkedLinked via unknownConstitutional methods for alignment →
- LinkedLinked via unknownLearning Complementary Action Modeling from Automotive Maintenance Instructions →
- PossiblePossibly related (embedding) · 46%modelscope/modelscope →
- PossiblePossibly related (embedding) · 51%Do Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied Reasoning →
- PossiblePossibly related (embedding) · 50%Bridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning Alignment →
- PossiblePossibly related (embedding) · 47%How Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs? →
- PossiblePossibly related (embedding) · 50%A Unified Moral-Value Dataset for Instruction Tuning →
Related to (incoming)
paperConstitutional methods for alignmentpaperLearning Complementary Action Modeling from Automotive Maintenance Instructionsrepomodelscope/modelscopepaperDo Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied ReasoningpaperBridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning AlignmentpaperHow Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?paperA Unified Moral-Value Dataset for Instruction TuningpaperBeyond Sycophancy: Structured Resistance and Compliance in LLM Moral Reasoning
Covers (incoming)
Related across the graph
paperConstitutional methods for alignmentpaperLearning Complementary Action Modeling from Automotive Maintenance InstructionspaperDo Vision-Language-Action Models Mean What They Say? On the Role of Faithfulness in Embodied ReasoningpaperA Unified Moral-Value Dataset for Instruction Tuningrepomodelscope/modelscopepaperHow Does Alignment Tuning Shape Representations of Sycophancy and Related Cue-Induced Biases in LLMs?paperBeyond Sycophancy: Structured Resistance and Compliance in LLM Moral ReasoningpaperBridging Physical Reasoning and Task Generalization via Visual Action Outcome Reasoning AlignmentnewsI made a quiz that tells you which LLM you align with most, based on personality and values research across 15 models [R]