Natural Language Autoencoders | Anthropic | Research | TechiSeek

AddAdd your Listings

Research

Natural Language Autoencoders

AI models like Claude talk in words but think in numbers. In this study, we train Claude to translate its thoughts into human-readable text.

Research Anthropic

ResearchAuditing language models for hidden objectivesAnthropic ResearchForecasting rare language model behaviorsAnthropic ResearchDecomposing language models into componentsAnthropic ResearchSycophancy to subterfuge: Investigating reward tampering in language modelsAnthropic ResearchFrom shortcuts to sabotage: natural emergent misalignment from reward hackingAnthropic ResearchConstitutional Classifiers: Defending against universal jailbreaksAnthropic Products & ServicesPre-recorded Speech-to-Text API | AssemblyAIAssemblyai ResearchThe Art of Asking: Multilingual Prompt Optimization for Synthetic DataCohere

ResearchBAM! Just Like That: Simple and Efficient Parameter Upcycling for Mixture of ExpertsCohere ResearchElo Uncovered: Robustness and Best Practices in Language Model EvaluationCohere ResearchNo News is Good News: A Critique of the One Billion Word BenchmarkCohere ResearchFrom One to Many: Expanding the Scope of Toxicity Mitigation in Language ModelsCohere BlogsUnlocking the potential of vision language models on satellite imagery through fine-tuningMistral BlogsHoneycomb Launches Integration With the Anthropic Usage and Cost APIHoneycomb ResearchUnderstanding and Mitigating Language Confusion in LLMsCohere ResearchContrastive Policy Gradient: Aligning LLMs on sequence-level scores in a supervised-friendly fashionCohere ResearchInvestigating Continual Pretraining in Large Language Models: Insights and ImplicationsCohere ResearchEAGER: Entropy-Aware Generation for Adaptive Inference-Time ScalingCohere ResearchScalable Data Ablation Approximations for Language Models through Modular Training and MergingCohere

BlogsI made a policy engine think it was in productionCncf